Skip to content

Optimize single-column dictionary grouping - #27272

Open
Rich-T-kid wants to merge 3 commits into
apache:mainfrom
Rich-T-kid:rich-t-kid/single-dictionary-group-values
Open

Rich-T-kid wants to merge 3 commits into
apache:mainfrom
Rich-T-kid:rich-t-kid/single-dictionary-group-values

Conversation

@Rich-T-kid

@Rich-T-kid Rich-T-kid commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

What does this PR do?

Adds a dedicated single-column dictionary grouping implementation. It resolves referenced dictionary values through the existing single-column value implementation, caches their group ids, and maps raw dictionary keys directly to those groups. Multi-column grouping is unchanged.

Related to #24822.

Target workload

This targets aggregates with one dictionary-encoded grouping expression, such as GROUP BY dictionary_column and COUNT(DISTINCT dictionary_column). It is most useful for low- and medium-cardinality strings where many rows reuse dictionary values. It does not affect scans, filters, or multi-column grouping.

Examples in ClickBench are Q5 (COUNT(DISTINCT SearchPhrase)), Q33 (GROUP BY URL), and Q34, where the constant grouping expression leaves URL as the only data grouping key.

Benchmarks

Bot results on f6ce57135, with RLE dictionary reads enabled for the query suites:

  • dictionary_group_values: all 35 cases improved by 1.27-4.10x.
  • TPC-H: all 22 queries unchanged.
  • TPC-DS: the first run flagged Q72 as 1.06x slower; the repeat instead measured Q72 as 1.09x faster and flagged Q28 as 1.07x slower, so no slowdown reproduced.
  • Partitioned ClickBench: 4 queries improved by 1.07-1.12x, 39 unchanged, none slower.

@github-actions github-actions Bot added the physical-plan Changes to the physical-plan crate label Nov 9, 2026
@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark tpch tpcds clickbench_partitioned

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750467359-4712-xgrb6 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpch

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750467359-4713-8j7jz 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpcds

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750467359-4714-dnmwg 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark clickbench_partitioned

Results will be posted here when complete


File an issue against this benchmark runner

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark dictionary_group_values

@adriangbot

Copy link
Copy Markdown

Benchmark for this request failed before finishing (Kubernetes reason: BackoffLimitExceeded).

Benchmarks requested: dictionary_group_values

Runner log (last 40 lines)
2026-11-09T20:22:46.924966Z  INFO runner starting benchmark runner bench_type=Datafusion, pr_url=https://github.com/apache/datafusion/pull/27272, benchmarks=dictionary_group_values
2026-11-09T20:22:46.987342Z  INFO benchmark_controller::runner::bench_datafusion === Cloning PR branch ===
2026-11-09T20:22:46.987365Z  INFO benchmark_controller::runner::shell running command cmd=git, args=["clone", "--depth=200", "https://github.com/apache/datafusion.git", "/workspace/datafusion-branch"], cwd="/"
2026-11-09T20:22:57.004004Z  INFO benchmark_controller::runner::shell running command cmd=git, args=["fetch", "origin", "refs/pull/27272/head:rich-t-kid/single-dictionary-group-values", "main"], cwd="/workspace/datafusion-branch"
2026-11-09T20:23:02.006585Z  INFO benchmark_controller::runner::shell running command cmd=git, args=["checkout", "rich-t-kid/single-dictionary-group-values"], cwd="/workspace/datafusion-branch"
2026-11-09T20:23:07.009053Z  INFO benchmark_controller::runner::shell running command cmd=git, args=["merge-base", "HEAD", "origin/main"], cwd="/workspace/datafusion-branch"
2026-11-09T20:23:12.011368Z  INFO benchmark_controller::runner::bench_datafusion === Cloning merge-base ===
2026-11-09T20:23:12.011384Z  INFO benchmark_controller::runner::shell running command cmd=git, args=["clone", "--depth=200", "https://github.com/apache/datafusion.git", "/workspace/datafusion-base"], cwd="/"
2026-11-09T20:25:27.041841Z ERROR runner benchmark failed error=git clone: git clone --depth=200 https://github.com/apache/datafusion.git /workspace/datafusion-base exited with code 128
stdout:

stderr:
Cloning into '/workspace/datafusion-base'...
fatal: unable to access 'https://github.com/apache/datafusion.git/': Failed to connect to github.com port 443 after 134061 ms: Couldn't connect to server
Kubernetes message
Job has reached the specified backoff limit

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750562195-4716-vf47f 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark dictionary_group_values

Results will be posted here when complete


File an issue against this benchmark runner

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark tpch tpcds clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750646201-4719-np9ml 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750646201-4718-22tnx 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6750646201-4717-hfhqc 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

Ignore the earlier three-suite run; it did not enable RLE dictionary reads. The rerun above does.

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpch
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 41.08 ms │                                  40.75 ms │ no change │
│ QQuery 2  │ 18.63 ms │                                  18.40 ms │ no change │
│ QQuery 3  │ 28.55 ms │                                  28.47 ms │ no change │
│ QQuery 4  │ 17.38 ms │                                  17.28 ms │ no change │
│ QQuery 5  │ 35.83 ms │                                  35.55 ms │ no change │
│ QQuery 6  │ 13.94 ms │                                  13.83 ms │ no change │
│ QQuery 7  │ 41.08 ms │                                  41.72 ms │ no change │
│ QQuery 8  │ 40.66 ms │                                  40.63 ms │ no change │
│ QQuery 9  │ 49.54 ms │                                  50.07 ms │ no change │
│ QQuery 10 │ 40.64 ms │                                  40.97 ms │ no change │
│ QQuery 11 │ 13.72 ms │                                  13.77 ms │ no change │
│ QQuery 12 │ 18.36 ms │                                  17.98 ms │ no change │
│ QQuery 13 │ 40.14 ms │                                  40.46 ms │ no change │
│ QQuery 14 │ 24.46 ms │                                  24.48 ms │ no change │
│ QQuery 15 │ 31.06 ms │                                  31.14 ms │ no change │
│ QQuery 16 │ 13.65 ms │                                  13.61 ms │ no change │
│ QQuery 17 │ 69.62 ms │                                  70.07 ms │ no change │
│ QQuery 18 │ 42.12 ms │                                  41.62 ms │ no change │
│ QQuery 19 │ 32.10 ms │                                  32.04 ms │ no change │
│ QQuery 20 │ 31.75 ms │                                  31.96 ms │ no change │
│ QQuery 21 │ 55.77 ms │                                  55.67 ms │ no change │
│ QQuery 22 │ 14.18 ms │                                  14.16 ms │ no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 714.25ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 714.61ms │
│ Average Time (HEAD)                                      │  32.47ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  32.48ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 41.08 / 43.25 ±2.42 / 47.76 ms │            40.75 / 41.82 ±1.03 / 43.27 ms │ no change │
│ QQuery 2  │ 18.63 / 18.81 ±0.14 / 19.01 ms │            18.40 / 18.58 ±0.17 / 18.90 ms │ no change │
│ QQuery 3  │ 28.55 / 28.73 ±0.21 / 29.02 ms │            28.47 / 28.66 ±0.13 / 28.82 ms │ no change │
│ QQuery 4  │ 17.38 / 17.55 ±0.15 / 17.83 ms │            17.28 / 17.40 ±0.08 / 17.53 ms │ no change │
│ QQuery 5  │ 35.83 / 36.11 ±0.34 / 36.71 ms │            35.55 / 35.93 ±0.24 / 36.21 ms │ no change │
│ QQuery 6  │ 13.94 / 14.21 ±0.37 / 14.94 ms │            13.83 / 13.91 ±0.07 / 14.00 ms │ no change │
│ QQuery 7  │ 41.08 / 43.05 ±1.76 / 46.01 ms │            41.72 / 43.16 ±1.45 / 45.65 ms │ no change │
│ QQuery 8  │ 40.66 / 43.55 ±2.80 / 48.42 ms │            40.63 / 42.14 ±1.27 / 44.16 ms │ no change │
│ QQuery 9  │ 49.54 / 50.85 ±1.52 / 53.81 ms │            50.07 / 51.39 ±1.14 / 52.78 ms │ no change │
│ QQuery 10 │ 40.64 / 41.16 ±0.75 / 42.64 ms │            40.97 / 42.14 ±1.36 / 44.38 ms │ no change │
│ QQuery 11 │ 13.72 / 14.10 ±0.50 / 15.09 ms │            13.77 / 13.92 ±0.19 / 14.29 ms │ no change │
│ QQuery 12 │ 18.36 / 18.56 ±0.26 / 19.06 ms │            17.98 / 18.55 ±0.62 / 19.70 ms │ no change │
│ QQuery 13 │ 40.14 / 41.34 ±1.33 / 43.93 ms │            40.46 / 41.73 ±1.64 / 44.62 ms │ no change │
│ QQuery 14 │ 24.46 / 24.71 ±0.38 / 25.45 ms │            24.48 / 24.64 ±0.11 / 24.79 ms │ no change │
│ QQuery 15 │ 31.06 / 31.67 ±0.91 / 33.47 ms │            31.14 / 31.85 ±0.59 / 32.63 ms │ no change │
│ QQuery 16 │ 13.65 / 13.85 ±0.21 / 14.21 ms │            13.61 / 13.77 ±0.17 / 14.10 ms │ no change │
│ QQuery 17 │ 69.62 / 70.59 ±1.18 / 72.83 ms │            70.07 / 71.42 ±1.16 / 72.98 ms │ no change │
│ QQuery 18 │ 42.12 / 42.47 ±0.31 / 42.94 ms │            41.62 / 43.05 ±1.86 / 46.63 ms │ no change │
│ QQuery 19 │ 32.10 / 32.35 ±0.19 / 32.62 ms │            32.04 / 32.54 ±0.31 / 32.97 ms │ no change │
│ QQuery 20 │ 31.75 / 31.95 ±0.18 / 32.23 ms │            31.96 / 32.31 ±0.23 / 32.62 ms │ no change │
│ QQuery 21 │ 55.77 / 56.99 ±0.97 / 58.13 ms │            55.67 / 57.09 ±1.12 / 58.66 ms │ no change │
│ QQuery 22 │ 14.18 / 14.25 ±0.10 / 14.44 ms │            14.16 / 14.26 ±0.09 / 14.43 ms │ no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 730.12ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 730.24ms │
│ Average Time (HEAD)                                      │  33.19ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  33.19ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 497.5 MiB
CPU user 19.6s
CPU sys 1.7s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 487.7 MiB
CPU user 19.6s
CPU sys 1.7s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpcds
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.30 ms │                                   5.24 ms │     no change │
│ QQuery 2  │   79.17 ms │                                  79.02 ms │     no change │
│ QQuery 3  │   28.80 ms │                                  28.80 ms │     no change │
│ QQuery 4  │  435.29 ms │                                 438.80 ms │     no change │
│ QQuery 5  │   50.29 ms │                                  49.79 ms │     no change │
│ QQuery 6  │   35.32 ms │                                  35.18 ms │     no change │
│ QQuery 7  │   71.71 ms │                                  72.24 ms │     no change │
│ QQuery 8  │   36.13 ms │                                  36.25 ms │     no change │
│ QQuery 9  │   51.73 ms │                                  51.99 ms │     no change │
│ QQuery 10 │   59.69 ms │                                  60.10 ms │     no change │
│ QQuery 11 │  270.92 ms │                                 276.45 ms │     no change │
│ QQuery 12 │   28.25 ms │                                  28.83 ms │     no change │
│ QQuery 13 │  112.50 ms │                                 112.65 ms │     no change │
│ QQuery 14 │  371.53 ms │                                 374.94 ms │     no change │
│ QQuery 15 │   52.90 ms │                                  53.09 ms │     no change │
│ QQuery 16 │    5.93 ms │                                   5.96 ms │     no change │
│ QQuery 17 │   75.26 ms │                                  75.05 ms │     no change │
│ QQuery 18 │  103.61 ms │                                 104.23 ms │     no change │
│ QQuery 19 │   40.85 ms │                                  40.87 ms │     no change │
│ QQuery 20 │   35.50 ms │                                  35.14 ms │     no change │
│ QQuery 21 │   16.87 ms │                                  17.09 ms │     no change │
│ QQuery 22 │   63.06 ms │                                  61.93 ms │     no change │
│ QQuery 23 │  290.29 ms │                                 290.59 ms │     no change │
│ QQuery 24 │  189.77 ms │                                 187.57 ms │     no change │
│ QQuery 25 │  103.11 ms │                                 104.36 ms │     no change │
│ QQuery 26 │   45.47 ms │                                  44.63 ms │     no change │
│ QQuery 27 │    4.86 ms │                                   4.84 ms │     no change │
│ QQuery 28 │   61.63 ms │                                  56.43 ms │ +1.09x faster │
│ QQuery 29 │   90.76 ms │                                  91.30 ms │     no change │
│ QQuery 30 │   32.14 ms │                                  31.47 ms │     no change │
│ QQuery 31 │  107.17 ms │                                 105.88 ms │     no change │
│ QQuery 32 │   19.89 ms │                                  19.56 ms │     no change │
│ QQuery 33 │   36.83 ms │                                  36.26 ms │     no change │
│ QQuery 34 │    5.35 ms │                                   5.32 ms │     no change │
│ QQuery 35 │   71.28 ms │                                  70.11 ms │     no change │
│ QQuery 36 │    5.47 ms │                                   5.63 ms │     no change │
│ QQuery 37 │    6.90 ms │                                   6.84 ms │     no change │
│ QQuery 38 │   59.62 ms │                                  59.69 ms │     no change │
│ QQuery 39 │   70.98 ms │                                  71.76 ms │     no change │
│ QQuery 40 │   22.54 ms │                                  21.95 ms │     no change │
│ QQuery 41 │   12.51 ms │                                  12.27 ms │     no change │
│ QQuery 42 │   23.63 ms │                                  23.18 ms │     no change │
│ QQuery 43 │    4.39 ms │                                   4.38 ms │     no change │
│ QQuery 44 │    5.47 ms │                                   5.30 ms │     no change │
│ QQuery 45 │   35.50 ms │                                  35.37 ms │     no change │
│ QQuery 46 │   11.67 ms │                                  11.26 ms │     no change │
│ QQuery 47 │  217.43 ms │                                 214.56 ms │     no change │
│ QQuery 48 │   91.26 ms │                                  91.12 ms │     no change │
│ QQuery 49 │   68.37 ms │                                  68.41 ms │     no change │
│ QQuery 50 │   55.85 ms │                                  55.78 ms │     no change │
│ QQuery 51 │   93.66 ms │                                  93.15 ms │     no change │
│ QQuery 52 │   23.61 ms │                                  23.58 ms │     no change │
│ QQuery 53 │   29.13 ms │                                  28.99 ms │     no change │
│ QQuery 54 │   52.79 ms │                                  52.82 ms │     no change │
│ QQuery 55 │   23.61 ms │                                  23.53 ms │     no change │
│ QQuery 56 │   37.40 ms │                                  37.45 ms │     no change │
│ QQuery 57 │  166.92 ms │                                 164.56 ms │     no change │
│ QQuery 58 │  105.09 ms │                                 106.06 ms │     no change │
│ QQuery 59 │  113.27 ms │                                 113.20 ms │     no change │
│ QQuery 60 │   37.86 ms │                                  38.24 ms │     no change │
│ QQuery 61 │    8.83 ms │                                   8.86 ms │     no change │
│ QQuery 62 │   43.12 ms │                                  43.10 ms │     no change │
│ QQuery 63 │   28.88 ms │                                  29.39 ms │     no change │
│ QQuery 64 │  344.40 ms │                                 345.85 ms │     no change │
│ QQuery 65 │   74.04 ms │                                  73.96 ms │     no change │
│ QQuery 66 │   73.74 ms │                                  72.17 ms │     no change │
│ QQuery 67 │  252.53 ms │                                 243.28 ms │     no change │
│ QQuery 68 │    6.41 ms │                                   6.21 ms │     no change │
│ QQuery 69 │   55.76 ms │                                  54.89 ms │     no change │
│ QQuery 70 │  102.95 ms │                                 101.93 ms │     no change │
│ QQuery 71 │   34.67 ms │                                  34.65 ms │     no change │
│ QQuery 72 │ 1640.06 ms │                                1610.33 ms │     no change │
│ QQuery 73 │    5.07 ms │                                   5.12 ms │     no change │
│ QQuery 74 │  183.61 ms │                                 185.56 ms │     no change │
│ QQuery 75 │  136.56 ms │                                 135.88 ms │     no change │
│ QQuery 76 │   33.72 ms │                                  33.56 ms │     no change │
│ QQuery 77 │   58.58 ms │                                  59.14 ms │     no change │
│ QQuery 78 │  263.29 ms │                                 265.41 ms │     no change │
│ QQuery 79 │   66.14 ms │                                  66.69 ms │     no change │
│ QQuery 80 │   93.29 ms │                                  93.61 ms │     no change │
│ QQuery 81 │   25.22 ms │                                  25.32 ms │     no change │
│ QQuery 82 │   16.18 ms │                                  16.37 ms │     no change │
│ QQuery 83 │   32.06 ms │                                  32.63 ms │     no change │
│ QQuery 84 │   28.97 ms │                                  29.83 ms │     no change │
│ QQuery 85 │   96.27 ms │                                  97.36 ms │     no change │
│ QQuery 86 │   25.20 ms │                                  25.17 ms │     no change │
│ QQuery 87 │   60.89 ms │                                  60.93 ms │     no change │
│ QQuery 88 │   59.01 ms │                                  60.32 ms │     no change │
│ QQuery 89 │   35.33 ms │                                  34.60 ms │     no change │
│ QQuery 90 │   16.00 ms │                                  16.76 ms │     no change │
│ QQuery 91 │   40.76 ms │                                  40.94 ms │     no change │
│ QQuery 92 │   29.18 ms │                                  29.39 ms │     no change │
│ QQuery 93 │   48.48 ms │                                  49.33 ms │     no change │
│ QQuery 94 │   37.94 ms │                                  38.74 ms │     no change │
│ QQuery 95 │   79.72 ms │                                  79.77 ms │     no change │
│ QQuery 96 │   23.96 ms │                                  23.40 ms │     no change │
│ QQuery 97 │   50.78 ms │                                  50.43 ms │     no change │
│ QQuery 98 │   42.74 ms │                                  42.63 ms │     no change │
│ QQuery 99 │   64.31 ms │                                  64.17 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 8782.71ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 8748.73ms │
│ Average Time (HEAD)                                      │   88.71ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   88.37ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │         0 │
│ Queries with No Change                                   │        98 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.30 / 5.88 ±1.08 / 8.03 ms │               5.24 / 5.77 ±0.92 / 7.61 ms │     no change │
│ QQuery 2  │        79.17 / 79.51 ±0.27 / 79.85 ms │            79.02 / 79.44 ±0.29 / 79.79 ms │     no change │
│ QQuery 3  │        28.80 / 29.07 ±0.16 / 29.24 ms │            28.80 / 29.02 ±0.20 / 29.32 ms │     no change │
│ QQuery 4  │     435.29 / 445.90 ±6.68 / 453.25 ms │         438.80 / 445.17 ±3.47 / 448.26 ms │     no change │
│ QQuery 5  │        50.29 / 52.01 ±2.29 / 56.24 ms │            49.79 / 51.87 ±3.67 / 59.19 ms │     no change │
│ QQuery 6  │        35.32 / 36.01 ±0.46 / 36.72 ms │            35.18 / 35.78 ±0.53 / 36.76 ms │     no change │
│ QQuery 7  │        71.71 / 72.27 ±0.54 / 73.03 ms │            72.24 / 72.80 ±0.48 / 73.43 ms │     no change │
│ QQuery 8  │        36.13 / 36.47 ±0.34 / 37.06 ms │            36.25 / 36.75 ±0.50 / 37.59 ms │     no change │
│ QQuery 9  │        51.73 / 56.22 ±4.71 / 64.09 ms │            51.99 / 53.26 ±1.78 / 56.68 ms │ +1.06x faster │
│ QQuery 10 │        59.69 / 60.39 ±0.47 / 61.14 ms │            60.10 / 60.68 ±0.50 / 61.28 ms │     no change │
│ QQuery 11 │     270.92 / 281.20 ±7.28 / 290.60 ms │        276.45 / 283.20 ±10.50 / 303.82 ms │     no change │
│ QQuery 12 │        28.25 / 28.81 ±0.34 / 29.23 ms │            28.83 / 29.19 ±0.35 / 29.81 ms │     no change │
│ QQuery 13 │     112.50 / 113.17 ±0.60 / 114.10 ms │         112.65 / 114.31 ±1.42 / 116.61 ms │     no change │
│ QQuery 14 │     371.53 / 376.46 ±2.57 / 378.18 ms │         374.94 / 377.40 ±2.19 / 381.51 ms │     no change │
│ QQuery 15 │        52.90 / 54.46 ±1.85 / 57.99 ms │            53.09 / 54.25 ±0.67 / 55.13 ms │     no change │
│ QQuery 16 │           5.93 / 6.08 ±0.20 / 6.47 ms │               5.96 / 6.15 ±0.23 / 6.59 ms │     no change │
│ QQuery 17 │        75.26 / 76.07 ±0.61 / 76.83 ms │            75.05 / 76.90 ±3.12 / 83.13 ms │     no change │
│ QQuery 18 │     103.61 / 104.86 ±1.30 / 107.31 ms │         104.23 / 104.87 ±0.55 / 105.68 ms │     no change │
│ QQuery 19 │        40.85 / 41.36 ±0.44 / 42.05 ms │            40.87 / 41.29 ±0.41 / 41.98 ms │     no change │
│ QQuery 20 │        35.50 / 36.08 ±0.39 / 36.53 ms │            35.14 / 35.89 ±0.57 / 36.62 ms │     no change │
│ QQuery 21 │        16.87 / 17.16 ±0.22 / 17.49 ms │            17.09 / 18.19 ±1.01 / 19.82 ms │  1.06x slower │
│ QQuery 22 │        63.06 / 65.35 ±3.96 / 73.26 ms │            61.93 / 63.71 ±1.20 / 65.04 ms │     no change │
│ QQuery 23 │     290.29 / 295.05 ±4.04 / 302.20 ms │         290.59 / 295.33 ±4.83 / 303.28 ms │     no change │
│ QQuery 24 │     189.77 / 192.50 ±1.57 / 194.37 ms │         187.57 / 191.44 ±3.96 / 198.48 ms │     no change │
│ QQuery 25 │     103.11 / 104.72 ±1.56 / 107.64 ms │         104.36 / 106.94 ±2.39 / 110.90 ms │     no change │
│ QQuery 26 │        45.47 / 45.78 ±0.31 / 46.31 ms │            44.63 / 45.08 ±0.34 / 45.58 ms │     no change │
│ QQuery 27 │          4.86 / 6.14 ±2.31 / 10.75 ms │               4.84 / 5.05 ±0.17 / 5.35 ms │ +1.22x faster │
│ QQuery 28 │        61.63 / 61.91 ±0.26 / 62.25 ms │            56.43 / 60.59 ±2.16 / 62.25 ms │     no change │
│ QQuery 29 │        90.76 / 92.42 ±1.60 / 95.35 ms │            91.30 / 93.20 ±1.69 / 96.02 ms │     no change │
│ QQuery 30 │        32.14 / 32.32 ±0.15 / 32.49 ms │            31.47 / 31.95 ±0.26 / 32.28 ms │     no change │
│ QQuery 31 │     107.17 / 108.45 ±2.00 / 112.43 ms │         105.88 / 107.01 ±1.41 / 109.75 ms │     no change │
│ QQuery 32 │        19.89 / 20.50 ±0.35 / 20.92 ms │            19.56 / 19.95 ±0.31 / 20.35 ms │     no change │
│ QQuery 33 │        36.83 / 37.99 ±0.98 / 39.72 ms │            36.26 / 36.76 ±0.64 / 37.92 ms │     no change │
│ QQuery 34 │           5.35 / 5.51 ±0.20 / 5.89 ms │               5.32 / 5.49 ±0.24 / 5.94 ms │     no change │
│ QQuery 35 │        71.28 / 73.06 ±2.38 / 77.63 ms │            70.11 / 70.71 ±0.50 / 71.61 ms │     no change │
│ QQuery 36 │           5.47 / 5.62 ±0.24 / 6.10 ms │               5.63 / 5.79 ±0.24 / 6.27 ms │     no change │
│ QQuery 37 │           6.90 / 7.02 ±0.13 / 7.27 ms │               6.84 / 6.97 ±0.08 / 7.10 ms │     no change │
│ QQuery 38 │        59.62 / 60.77 ±0.98 / 62.40 ms │            59.69 / 60.71 ±1.32 / 63.29 ms │     no change │
│ QQuery 39 │        70.98 / 72.55 ±1.46 / 75.27 ms │            71.76 / 72.03 ±0.29 / 72.55 ms │     no change │
│ QQuery 40 │        22.54 / 24.11 ±2.73 / 29.55 ms │            21.95 / 22.09 ±0.13 / 22.33 ms │ +1.09x faster │
│ QQuery 41 │        12.51 / 12.77 ±0.20 / 13.08 ms │            12.27 / 12.45 ±0.13 / 12.64 ms │     no change │
│ QQuery 42 │        23.63 / 23.95 ±0.25 / 24.21 ms │            23.18 / 23.80 ±0.40 / 24.38 ms │     no change │
│ QQuery 43 │           4.39 / 4.54 ±0.19 / 4.92 ms │               4.38 / 4.49 ±0.17 / 4.83 ms │     no change │
│ QQuery 44 │           5.47 / 5.54 ±0.10 / 5.73 ms │               5.30 / 5.38 ±0.07 / 5.48 ms │     no change │
│ QQuery 45 │        35.50 / 36.78 ±0.80 / 37.82 ms │            35.37 / 36.06 ±0.38 / 36.47 ms │     no change │
│ QQuery 46 │        11.67 / 12.15 ±0.36 / 12.59 ms │            11.26 / 11.96 ±0.67 / 13.25 ms │     no change │
│ QQuery 47 │     217.43 / 221.68 ±3.46 / 227.07 ms │         214.56 / 219.35 ±4.91 / 225.32 ms │     no change │
│ QQuery 48 │        91.26 / 93.12 ±2.93 / 98.95 ms │            91.12 / 91.62 ±0.58 / 92.65 ms │     no change │
│ QQuery 49 │        68.37 / 69.24 ±0.64 / 70.29 ms │            68.41 / 69.13 ±0.38 / 69.48 ms │     no change │
│ QQuery 50 │        55.85 / 56.54 ±0.89 / 58.14 ms │            55.78 / 58.89 ±2.82 / 62.86 ms │     no change │
│ QQuery 51 │        93.66 / 95.15 ±1.88 / 98.81 ms │            93.15 / 93.94 ±0.79 / 95.27 ms │     no change │
│ QQuery 52 │        23.61 / 24.03 ±0.35 / 24.66 ms │            23.58 / 23.86 ±0.17 / 24.04 ms │     no change │
│ QQuery 53 │        29.13 / 29.28 ±0.11 / 29.46 ms │            28.99 / 29.14 ±0.12 / 29.35 ms │     no change │
│ QQuery 54 │        52.79 / 53.63 ±0.89 / 55.27 ms │            52.82 / 54.16 ±1.57 / 57.21 ms │     no change │
│ QQuery 55 │        23.61 / 23.96 ±0.39 / 24.72 ms │            23.53 / 23.92 ±0.40 / 24.66 ms │     no change │
│ QQuery 56 │        37.40 / 37.92 ±0.47 / 38.81 ms │            37.45 / 37.83 ±0.27 / 38.21 ms │     no change │
│ QQuery 57 │     166.92 / 167.88 ±0.72 / 168.85 ms │         164.56 / 168.98 ±2.57 / 172.03 ms │     no change │
│ QQuery 58 │     105.09 / 106.73 ±1.70 / 109.72 ms │         106.06 / 106.78 ±0.52 / 107.48 ms │     no change │
│ QQuery 59 │     113.27 / 114.20 ±0.77 / 115.54 ms │         113.20 / 115.57 ±2.99 / 121.45 ms │     no change │
│ QQuery 60 │        37.86 / 38.81 ±0.92 / 40.57 ms │            38.24 / 38.91 ±0.44 / 39.53 ms │     no change │
│ QQuery 61 │           8.83 / 8.99 ±0.21 / 9.41 ms │               8.86 / 9.04 ±0.20 / 9.42 ms │     no change │
│ QQuery 62 │        43.12 / 43.38 ±0.32 / 44.01 ms │            43.10 / 43.34 ±0.13 / 43.45 ms │     no change │
│ QQuery 63 │        28.88 / 29.16 ±0.20 / 29.39 ms │            29.39 / 32.03 ±4.44 / 40.82 ms │  1.10x slower │
│ QQuery 64 │     344.40 / 347.36 ±2.79 / 350.98 ms │         345.85 / 350.93 ±3.67 / 357.18 ms │     no change │
│ QQuery 65 │        74.04 / 74.91 ±0.80 / 76.29 ms │            73.96 / 77.41 ±4.72 / 86.70 ms │     no change │
│ QQuery 66 │        73.74 / 76.07 ±2.48 / 79.60 ms │            72.17 / 72.76 ±0.60 / 73.57 ms │     no change │
│ QQuery 67 │     252.53 / 258.62 ±3.51 / 262.51 ms │         243.28 / 249.03 ±5.04 / 255.69 ms │     no change │
│ QQuery 68 │           6.41 / 6.57 ±0.18 / 6.92 ms │               6.21 / 6.52 ±0.32 / 7.14 ms │     no change │
│ QQuery 69 │        55.76 / 56.49 ±0.70 / 57.51 ms │            54.89 / 59.22 ±4.53 / 67.53 ms │     no change │
│ QQuery 70 │     102.95 / 108.10 ±4.25 / 115.87 ms │         101.93 / 103.51 ±2.20 / 107.87 ms │     no change │
│ QQuery 71 │        34.67 / 35.33 ±0.37 / 35.77 ms │            34.65 / 35.48 ±0.63 / 36.48 ms │     no change │
│ QQuery 72 │ 1640.06 / 1719.87 ±54.38 / 1780.44 ms │     1610.33 / 1676.19 ±45.45 / 1728.97 ms │     no change │
│ QQuery 73 │           5.07 / 5.25 ±0.23 / 5.70 ms │               5.12 / 5.27 ±0.21 / 5.68 ms │     no change │
│ QQuery 74 │     183.61 / 186.13 ±1.32 / 187.31 ms │         185.56 / 190.28 ±3.11 / 194.70 ms │     no change │
│ QQuery 75 │     136.56 / 139.03 ±3.75 / 146.49 ms │         135.88 / 136.99 ±0.76 / 137.92 ms │     no change │
│ QQuery 76 │        33.72 / 33.90 ±0.18 / 34.21 ms │            33.56 / 35.63 ±3.33 / 42.27 ms │  1.05x slower │
│ QQuery 77 │        58.58 / 60.12 ±1.85 / 63.61 ms │            59.14 / 60.80 ±2.58 / 65.92 ms │     no change │
│ QQuery 78 │    263.29 / 278.75 ±17.52 / 303.97 ms │        265.41 / 277.25 ±12.47 / 292.95 ms │     no change │
│ QQuery 79 │        66.14 / 66.42 ±0.27 / 66.90 ms │            66.69 / 69.49 ±4.87 / 79.21 ms │     no change │
│ QQuery 80 │       93.29 / 96.56 ±4.84 / 106.17 ms │            93.61 / 94.42 ±1.09 / 96.55 ms │     no change │
│ QQuery 81 │        25.22 / 25.43 ±0.23 / 25.84 ms │            25.32 / 25.81 ±0.29 / 26.12 ms │     no change │
│ QQuery 82 │        16.18 / 16.55 ±0.19 / 16.75 ms │            16.37 / 17.30 ±1.48 / 20.24 ms │     no change │
│ QQuery 83 │        32.06 / 32.19 ±0.07 / 32.26 ms │            32.63 / 32.97 ±0.25 / 33.26 ms │     no change │
│ QQuery 84 │        28.97 / 29.09 ±0.08 / 29.20 ms │            29.83 / 31.95 ±1.98 / 34.33 ms │  1.10x slower │
│ QQuery 85 │       96.27 / 98.92 ±3.35 / 105.23 ms │           97.36 / 98.42 ±1.02 / 100.10 ms │     no change │
│ QQuery 86 │        25.20 / 25.54 ±0.25 / 25.94 ms │            25.17 / 25.64 ±0.38 / 26.29 ms │     no change │
│ QQuery 87 │        60.89 / 61.16 ±0.23 / 61.57 ms │            60.93 / 62.12 ±0.72 / 62.93 ms │     no change │
│ QQuery 88 │        59.01 / 60.00 ±0.58 / 60.65 ms │            60.32 / 61.96 ±1.54 / 63.91 ms │     no change │
│ QQuery 89 │        35.33 / 36.39 ±1.48 / 39.27 ms │            34.60 / 35.94 ±0.88 / 37.26 ms │     no change │
│ QQuery 90 │        16.00 / 16.31 ±0.28 / 16.74 ms │            16.76 / 16.99 ±0.11 / 17.07 ms │     no change │
│ QQuery 91 │        40.76 / 41.21 ±0.33 / 41.58 ms │            40.94 / 41.24 ±0.21 / 41.59 ms │     no change │
│ QQuery 92 │        29.18 / 29.56 ±0.35 / 30.12 ms │            29.39 / 29.54 ±0.10 / 29.67 ms │     no change │
│ QQuery 93 │        48.48 / 48.98 ±0.49 / 49.61 ms │            49.33 / 52.48 ±3.86 / 60.02 ms │  1.07x slower │
│ QQuery 94 │        37.94 / 39.05 ±1.64 / 42.26 ms │            38.74 / 39.49 ±0.48 / 40.22 ms │     no change │
│ QQuery 95 │        79.72 / 80.86 ±0.88 / 82.07 ms │            79.77 / 80.29 ±0.39 / 80.79 ms │     no change │
│ QQuery 96 │        23.96 / 24.26 ±0.35 / 24.95 ms │            23.40 / 23.73 ±0.20 / 23.94 ms │     no change │
│ QQuery 97 │        50.78 / 52.05 ±1.12 / 53.94 ms │            50.43 / 52.32 ±2.85 / 57.95 ms │     no change │
│ QQuery 98 │        42.74 / 43.65 ±0.59 / 44.38 ms │            42.63 / 44.35 ±0.89 / 45.08 ms │     no change │
│ QQuery 99 │        64.31 / 64.83 ±0.45 / 65.52 ms │            64.17 / 64.54 ±0.30 / 64.93 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9006.13ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 8961.84ms │
│ Average Time (HEAD)                                      │   90.97ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   90.52ms │
│ Queries Faster                                           │         3 │
│ Queries Slower                                           │         5 │
│ Queries with No Change                                   │        91 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.3 GiB
Avg memory 1.5 GiB
CPU user 188.4s
CPU sys 5.5s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.0 GiB
Avg memory 1.4 GiB
CPU user 189.1s
CPU sys 5.6s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 40.56 ms │                                  39.79 ms │ no change │
│ QQuery 2  │ 19.36 ms │                                  19.32 ms │ no change │
│ QQuery 3  │ 28.92 ms │                                  28.95 ms │ no change │
│ QQuery 4  │ 17.68 ms │                                  17.37 ms │ no change │
│ QQuery 5  │ 36.54 ms │                                  36.39 ms │ no change │
│ QQuery 6  │ 13.96 ms │                                  14.02 ms │ no change │
│ QQuery 7  │ 41.50 ms │                                  42.73 ms │ no change │
│ QQuery 8  │ 41.16 ms │                                  41.22 ms │ no change │
│ QQuery 9  │ 50.30 ms │                                  50.78 ms │ no change │
│ QQuery 10 │ 60.21 ms │                                  59.72 ms │ no change │
│ QQuery 11 │ 14.34 ms │                                  14.32 ms │ no change │
│ QQuery 12 │ 21.35 ms │                                  21.17 ms │ no change │
│ QQuery 13 │ 48.01 ms │                                  47.67 ms │ no change │
│ QQuery 14 │ 24.45 ms │                                  24.37 ms │ no change │
│ QQuery 15 │ 31.70 ms │                                  31.91 ms │ no change │
│ QQuery 16 │ 13.80 ms │                                  13.90 ms │ no change │
│ QQuery 17 │ 70.83 ms │                                  71.71 ms │ no change │
│ QQuery 18 │ 67.99 ms │                                  66.01 ms │ no change │
│ QQuery 19 │ 29.79 ms │                                  29.86 ms │ no change │
│ QQuery 20 │ 34.70 ms │                                  34.58 ms │ no change │
│ QQuery 21 │ 55.13 ms │                                  54.21 ms │ no change │
│ QQuery 22 │ 16.03 ms │                                  16.10 ms │ no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 778.32ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 776.10ms │
│ Average Time (HEAD)                                      │  35.38ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  35.28ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 40.56 / 42.14 ±1.28 / 43.55 ms │            39.79 / 40.48 ±1.09 / 42.65 ms │ no change │
│ QQuery 2  │ 19.36 / 19.90 ±0.66 / 21.15 ms │            19.32 / 19.58 ±0.22 / 19.98 ms │ no change │
│ QQuery 3  │ 28.92 / 29.33 ±0.26 / 29.65 ms │            28.95 / 29.67 ±1.31 / 32.29 ms │ no change │
│ QQuery 4  │ 17.68 / 17.82 ±0.13 / 18.06 ms │            17.37 / 17.49 ±0.14 / 17.70 ms │ no change │
│ QQuery 5  │ 36.54 / 36.72 ±0.14 / 36.90 ms │            36.39 / 36.66 ±0.14 / 36.82 ms │ no change │
│ QQuery 6  │ 13.96 / 14.40 ±0.40 / 15.05 ms │            14.02 / 14.32 ±0.33 / 14.94 ms │ no change │
│ QQuery 7  │ 41.50 / 43.29 ±1.52 / 45.84 ms │            42.73 / 44.16 ±2.58 / 49.33 ms │ no change │
│ QQuery 8  │ 41.16 / 41.49 ±0.24 / 41.81 ms │            41.22 / 42.17 ±0.99 / 44.04 ms │ no change │
│ QQuery 9  │ 50.30 / 51.09 ±0.47 / 51.71 ms │            50.78 / 51.99 ±1.77 / 55.46 ms │ no change │
│ QQuery 10 │ 60.21 / 60.55 ±0.42 / 61.36 ms │            59.72 / 60.03 ±0.23 / 60.23 ms │ no change │
│ QQuery 11 │ 14.34 / 14.81 ±0.77 / 16.34 ms │            14.32 / 14.75 ±0.51 / 15.70 ms │ no change │
│ QQuery 12 │ 21.35 / 21.44 ±0.10 / 21.58 ms │            21.17 / 21.64 ±0.53 / 22.39 ms │ no change │
│ QQuery 13 │ 48.01 / 50.37 ±2.35 / 54.45 ms │            47.67 / 48.89 ±1.41 / 51.66 ms │ no change │
│ QQuery 14 │ 24.45 / 25.04 ±0.98 / 26.99 ms │            24.37 / 24.52 ±0.10 / 24.63 ms │ no change │
│ QQuery 15 │ 31.70 / 32.87 ±0.97 / 34.40 ms │            31.91 / 32.59 ±0.46 / 33.34 ms │ no change │
│ QQuery 16 │ 13.80 / 13.93 ±0.09 / 14.03 ms │            13.90 / 13.99 ±0.07 / 14.09 ms │ no change │
│ QQuery 17 │ 70.83 / 72.81 ±1.33 / 74.99 ms │            71.71 / 72.83 ±1.02 / 74.65 ms │ no change │
│ QQuery 18 │ 67.99 / 70.29 ±1.55 / 72.77 ms │            66.01 / 68.22 ±1.72 / 70.64 ms │ no change │
│ QQuery 19 │ 29.79 / 30.05 ±0.24 / 30.50 ms │            29.86 / 30.03 ±0.13 / 30.16 ms │ no change │
│ QQuery 20 │ 34.70 / 35.04 ±0.25 / 35.40 ms │            34.58 / 34.79 ±0.14 / 35.01 ms │ no change │
│ QQuery 21 │ 55.13 / 56.61 ±1.04 / 58.07 ms │            54.21 / 54.88 ±0.79 / 56.39 ms │ no change │
│ QQuery 22 │ 16.03 / 16.26 ±0.15 / 16.45 ms │            16.10 / 16.42 ±0.33 / 17.04 ms │ no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 796.24ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 790.11ms │
│ Average Time (HEAD)                                      │  36.19ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  35.91ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 464.8 MiB
CPU user 20.8s
CPU sys 1.7s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 483.7 MiB
CPU user 20.7s
CPU sys 1.7s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │    5.41 ms │                                   5.24 ms │ no change │
│ QQuery 2  │   75.48 ms │                                  75.99 ms │ no change │
│ QQuery 3  │   28.40 ms │                                  28.87 ms │ no change │
│ QQuery 4  │  566.45 ms │                                 568.22 ms │ no change │
│ QQuery 5  │   49.42 ms │                                  50.16 ms │ no change │
│ QQuery 6  │   32.90 ms │                                  33.15 ms │ no change │
│ QQuery 7  │   69.88 ms │                                  70.46 ms │ no change │
│ QQuery 8  │   38.28 ms │                                  38.40 ms │ no change │
│ QQuery 9  │   53.51 ms │                                  51.06 ms │ no change │
│ QQuery 10 │   58.82 ms │                                  58.91 ms │ no change │
│ QQuery 11 │  371.46 ms │                                 376.23 ms │ no change │
│ QQuery 12 │   25.71 ms │                                  25.65 ms │ no change │
│ QQuery 13 │  110.99 ms │                                 111.64 ms │ no change │
│ QQuery 14 │  373.96 ms │                                 375.15 ms │ no change │
│ QQuery 15 │   51.12 ms │                                  50.48 ms │ no change │
│ QQuery 16 │    5.87 ms │                                   5.80 ms │ no change │
│ QQuery 17 │   76.00 ms │                                  75.82 ms │ no change │
│ QQuery 18 │  102.49 ms │                                 102.57 ms │ no change │
│ QQuery 19 │   41.38 ms │                                  41.13 ms │ no change │
│ QQuery 20 │   30.59 ms │                                  30.99 ms │ no change │
│ QQuery 21 │   17.09 ms │                                  16.83 ms │ no change │
│ QQuery 22 │   68.67 ms │                                  68.99 ms │ no change │
│ QQuery 23 │  286.82 ms │                                 287.68 ms │ no change │
│ QQuery 24 │  187.95 ms │                                 190.62 ms │ no change │
│ QQuery 25 │  104.09 ms │                                 104.78 ms │ no change │
│ QQuery 26 │   44.07 ms │                                  44.23 ms │ no change │
│ QQuery 27 │    5.12 ms │                                   5.09 ms │ no change │
│ QQuery 28 │   56.11 ms │                                  56.37 ms │ no change │
│ QQuery 29 │   92.07 ms │                                  92.86 ms │ no change │
│ QQuery 30 │   35.51 ms │                                  36.31 ms │ no change │
│ QQuery 31 │  108.10 ms │                                 108.81 ms │ no change │
│ QQuery 32 │   19.92 ms │                                  19.76 ms │ no change │
│ QQuery 33 │   36.82 ms │                                  37.01 ms │ no change │
│ QQuery 34 │    5.63 ms │                                   5.76 ms │ no change │
│ QQuery 35 │   69.64 ms │                                  69.11 ms │ no change │
│ QQuery 36 │    5.80 ms │                                   5.67 ms │ no change │
│ QQuery 37 │    7.34 ms │                                   7.17 ms │ no change │
│ QQuery 38 │   63.35 ms │                                  63.73 ms │ no change │
│ QQuery 39 │   71.13 ms │                                  71.91 ms │ no change │
│ QQuery 40 │   22.28 ms │                                  22.56 ms │ no change │
│ QQuery 41 │   14.06 ms │                                  14.16 ms │ no change │
│ QQuery 42 │   23.49 ms │                                  24.05 ms │ no change │
│ QQuery 43 │    4.63 ms │                                   4.69 ms │ no change │
│ QQuery 44 │    5.45 ms │                                   5.46 ms │ no change │
│ QQuery 45 │   35.39 ms │                                  35.36 ms │ no change │
│ QQuery 46 │   11.17 ms │                                  11.46 ms │ no change │
│ QQuery 47 │  199.52 ms │                                 201.76 ms │ no change │
│ QQuery 48 │   90.15 ms │                                  90.58 ms │ no change │
│ QQuery 49 │   68.09 ms │                                  68.45 ms │ no change │
│ QQuery 50 │   56.79 ms │                                  57.26 ms │ no change │
│ QQuery 51 │   93.37 ms │                                  92.26 ms │ no change │
│ QQuery 52 │   24.10 ms │                                  24.07 ms │ no change │
│ QQuery 53 │   29.12 ms │                                  29.45 ms │ no change │
│ QQuery 54 │   53.19 ms │                                  53.51 ms │ no change │
│ QQuery 55 │   23.74 ms │                                  23.94 ms │ no change │
│ QQuery 56 │   37.52 ms │                                  38.03 ms │ no change │
│ QQuery 57 │  150.65 ms │                                 151.72 ms │ no change │
│ QQuery 58 │  102.06 ms │                                 102.73 ms │ no change │
│ QQuery 59 │  106.60 ms │                                 107.67 ms │ no change │
│ QQuery 60 │   38.14 ms │                                  37.81 ms │ no change │
│ QQuery 61 │    8.86 ms │                                   8.95 ms │ no change │
│ QQuery 62 │   37.57 ms │                                  37.50 ms │ no change │
│ QQuery 63 │   29.43 ms │                                  29.44 ms │ no change │
│ QQuery 64 │  351.89 ms │                                 355.07 ms │ no change │
│ QQuery 65 │   75.38 ms │                                  75.13 ms │ no change │
│ QQuery 66 │   71.15 ms │                                  71.39 ms │ no change │
│ QQuery 67 │  220.64 ms │                                 221.75 ms │ no change │
│ QQuery 68 │    6.42 ms │                                   6.46 ms │ no change │
│ QQuery 69 │   53.35 ms │                                  53.43 ms │ no change │
│ QQuery 70 │   95.48 ms │                                  96.59 ms │ no change │
│ QQuery 71 │   34.86 ms │                                  34.83 ms │ no change │
│ QQuery 72 │ 1571.87 ms │                                1564.71 ms │ no change │
│ QQuery 73 │    5.17 ms │                                   5.25 ms │ no change │
│ QQuery 74 │  242.94 ms │                                 239.77 ms │ no change │
│ QQuery 75 │  137.27 ms │                                 137.78 ms │ no change │
│ QQuery 76 │   34.37 ms │                                  34.39 ms │ no change │
│ QQuery 77 │   59.49 ms │                                  59.75 ms │ no change │
│ QQuery 78 │  257.70 ms │                                 260.85 ms │ no change │
│ QQuery 79 │   66.88 ms │                                  67.05 ms │ no change │
│ QQuery 80 │   92.34 ms │                                  93.46 ms │ no change │
│ QQuery 81 │   28.36 ms │                                  28.17 ms │ no change │
│ QQuery 82 │   17.44 ms │                                  17.04 ms │ no change │
│ QQuery 83 │   32.03 ms │                                  31.88 ms │ no change │
│ QQuery 84 │   30.45 ms │                                  30.47 ms │ no change │
│ QQuery 85 │   94.67 ms │                                  95.17 ms │ no change │
│ QQuery 86 │   24.01 ms │                                  23.92 ms │ no change │
│ QQuery 87 │   64.51 ms │                                  64.35 ms │ no change │
│ QQuery 88 │   59.06 ms │                                  60.56 ms │ no change │
│ QQuery 89 │   35.42 ms │                                  36.45 ms │ no change │
│ QQuery 90 │   16.01 ms │                                  16.41 ms │ no change │
│ QQuery 91 │   39.58 ms │                                  39.96 ms │ no change │
│ QQuery 92 │   28.84 ms │                                  28.94 ms │ no change │
│ QQuery 93 │   48.70 ms │                                  49.00 ms │ no change │
│ QQuery 94 │   37.73 ms │                                  38.29 ms │ no change │
│ QQuery 95 │   78.34 ms │                                  80.54 ms │ no change │
│ QQuery 96 │   23.44 ms │                                  23.72 ms │ no change │
│ QQuery 97 │   50.08 ms │                                  50.58 ms │ no change │
│ QQuery 98 │   37.34 ms │                                  38.60 ms │ no change │
│ QQuery 99 │   56.13 ms │                                  56.96 ms │ no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 8898.04ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 8930.17ms │
│ Average Time (HEAD)                                      │   89.88ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   90.20ms │
│ Queries Faster                                           │         0 │
│ Queries Slower                                           │         0 │
│ Queries with No Change                                   │        99 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.41 / 6.05 ±1.10 / 8.24 ms │               5.24 / 5.93 ±1.14 / 8.19 ms │     no change │
│ QQuery 2  │        75.48 / 76.06 ±0.43 / 76.80 ms │            75.99 / 76.60 ±0.33 / 76.96 ms │     no change │
│ QQuery 3  │        28.40 / 28.61 ±0.15 / 28.80 ms │            28.87 / 28.97 ±0.09 / 29.14 ms │     no change │
│ QQuery 4  │     566.45 / 569.84 ±2.22 / 572.77 ms │         568.22 / 572.78 ±2.97 / 576.22 ms │     no change │
│ QQuery 5  │        49.42 / 50.08 ±0.43 / 50.73 ms │            50.16 / 50.40 ±0.20 / 50.69 ms │     no change │
│ QQuery 6  │        32.90 / 33.31 ±0.30 / 33.78 ms │            33.15 / 33.61 ±0.25 / 33.89 ms │     no change │
│ QQuery 7  │        69.88 / 72.35 ±2.87 / 77.86 ms │            70.46 / 72.37 ±2.29 / 76.73 ms │     no change │
│ QQuery 8  │        38.28 / 39.07 ±0.85 / 40.60 ms │            38.40 / 38.96 ±0.42 / 39.68 ms │     no change │
│ QQuery 9  │        53.51 / 55.02 ±1.13 / 55.95 ms │            51.06 / 54.18 ±1.64 / 55.72 ms │     no change │
│ QQuery 10 │        58.82 / 59.33 ±0.38 / 59.97 ms │            58.91 / 59.29 ±0.50 / 60.25 ms │     no change │
│ QQuery 11 │     371.46 / 377.51 ±5.23 / 385.42 ms │         376.23 / 379.71 ±2.59 / 382.97 ms │     no change │
│ QQuery 12 │        25.71 / 25.82 ±0.09 / 25.97 ms │            25.65 / 25.93 ±0.16 / 26.13 ms │     no change │
│ QQuery 13 │     110.99 / 112.04 ±1.10 / 114.14 ms │         111.64 / 113.35 ±2.66 / 118.66 ms │     no change │
│ QQuery 14 │     373.96 / 375.76 ±1.93 / 379.01 ms │         375.15 / 376.68 ±1.66 / 379.52 ms │     no change │
│ QQuery 15 │        51.12 / 52.24 ±0.71 / 53.09 ms │            50.48 / 51.67 ±0.79 / 52.72 ms │     no change │
│ QQuery 16 │           5.87 / 6.02 ±0.17 / 6.32 ms │               5.80 / 5.96 ±0.20 / 6.36 ms │     no change │
│ QQuery 17 │        76.00 / 77.07 ±2.02 / 81.10 ms │            75.82 / 78.02 ±2.37 / 81.50 ms │     no change │
│ QQuery 18 │     102.49 / 103.65 ±1.11 / 105.50 ms │         102.57 / 103.64 ±0.73 / 104.50 ms │     no change │
│ QQuery 19 │        41.38 / 41.66 ±0.37 / 42.37 ms │            41.13 / 41.58 ±0.51 / 42.55 ms │     no change │
│ QQuery 20 │        30.59 / 30.96 ±0.20 / 31.18 ms │            30.99 / 31.42 ±0.25 / 31.74 ms │     no change │
│ QQuery 21 │        17.09 / 17.92 ±0.97 / 19.42 ms │            16.83 / 18.12 ±2.14 / 22.38 ms │     no change │
│ QQuery 22 │        68.67 / 69.31 ±0.55 / 70.31 ms │            68.99 / 69.66 ±0.60 / 70.71 ms │     no change │
│ QQuery 23 │     286.82 / 290.81 ±2.99 / 295.70 ms │         287.68 / 293.22 ±5.30 / 303.34 ms │     no change │
│ QQuery 24 │     187.95 / 190.46 ±3.26 / 196.28 ms │         190.62 / 194.37 ±4.60 / 203.36 ms │     no change │
│ QQuery 25 │     104.09 / 106.29 ±2.00 / 109.68 ms │         104.78 / 106.16 ±1.05 / 107.56 ms │     no change │
│ QQuery 26 │        44.07 / 44.28 ±0.12 / 44.42 ms │            44.23 / 44.55 ±0.46 / 45.44 ms │     no change │
│ QQuery 27 │           5.12 / 5.32 ±0.18 / 5.60 ms │               5.09 / 5.20 ±0.15 / 5.49 ms │     no change │
│ QQuery 28 │        56.11 / 60.49 ±2.22 / 62.09 ms │            56.37 / 58.50 ±2.21 / 61.30 ms │     no change │
│ QQuery 29 │        92.07 / 94.77 ±2.86 / 99.58 ms │           92.86 / 96.77 ±5.70 / 107.88 ms │     no change │
│ QQuery 30 │        35.51 / 35.73 ±0.19 / 36.05 ms │            36.31 / 36.64 ±0.34 / 37.24 ms │     no change │
│ QQuery 31 │     108.10 / 109.41 ±1.30 / 111.85 ms │         108.81 / 110.96 ±3.14 / 117.16 ms │     no change │
│ QQuery 32 │        19.92 / 20.24 ±0.26 / 20.69 ms │            19.76 / 20.33 ±0.46 / 21.00 ms │     no change │
│ QQuery 33 │        36.82 / 37.39 ±0.54 / 38.38 ms │            37.01 / 37.35 ±0.32 / 37.93 ms │     no change │
│ QQuery 34 │           5.63 / 5.79 ±0.21 / 6.18 ms │               5.76 / 5.94 ±0.24 / 6.41 ms │     no change │
│ QQuery 35 │        69.64 / 70.13 ±0.31 / 70.43 ms │            69.11 / 69.38 ±0.30 / 69.95 ms │     no change │
│ QQuery 36 │           5.80 / 6.00 ±0.23 / 6.45 ms │               5.67 / 5.82 ±0.24 / 6.29 ms │     no change │
│ QQuery 37 │           7.34 / 7.52 ±0.14 / 7.66 ms │               7.17 / 7.41 ±0.13 / 7.57 ms │     no change │
│ QQuery 38 │        63.35 / 65.24 ±2.50 / 70.11 ms │            63.73 / 64.93 ±1.19 / 67.04 ms │     no change │
│ QQuery 39 │        71.13 / 71.82 ±0.64 / 72.96 ms │            71.91 / 72.37 ±0.26 / 72.66 ms │     no change │
│ QQuery 40 │        22.28 / 22.64 ±0.32 / 23.12 ms │            22.56 / 22.83 ±0.22 / 23.10 ms │     no change │
│ QQuery 41 │        14.06 / 14.32 ±0.26 / 14.78 ms │            14.16 / 14.47 ±0.25 / 14.89 ms │     no change │
│ QQuery 42 │        23.49 / 23.88 ±0.35 / 24.36 ms │            24.05 / 24.37 ±0.25 / 24.71 ms │     no change │
│ QQuery 43 │           4.63 / 4.76 ±0.24 / 5.24 ms │               4.69 / 4.84 ±0.24 / 5.31 ms │     no change │
│ QQuery 44 │           5.45 / 5.52 ±0.04 / 5.57 ms │               5.46 / 5.57 ±0.11 / 5.77 ms │     no change │
│ QQuery 45 │        35.39 / 37.19 ±2.73 / 42.63 ms │            35.36 / 37.73 ±1.94 / 40.09 ms │     no change │
│ QQuery 46 │        11.17 / 11.53 ±0.50 / 12.52 ms │            11.46 / 11.70 ±0.26 / 12.17 ms │     no change │
│ QQuery 47 │     199.52 / 204.03 ±4.50 / 212.28 ms │         201.76 / 204.45 ±2.28 / 208.62 ms │     no change │
│ QQuery 48 │        90.15 / 90.68 ±0.32 / 91.10 ms │            90.58 / 91.14 ±0.59 / 92.19 ms │     no change │
│ QQuery 49 │        68.09 / 68.93 ±0.62 / 69.65 ms │            68.45 / 69.23 ±0.88 / 70.91 ms │     no change │
│ QQuery 50 │        56.79 / 60.40 ±5.23 / 70.66 ms │            57.26 / 60.18 ±3.35 / 66.35 ms │     no change │
│ QQuery 51 │        93.37 / 94.65 ±1.56 / 97.63 ms │            92.26 / 94.49 ±1.43 / 96.56 ms │     no change │
│ QQuery 52 │        24.10 / 24.48 ±0.38 / 25.19 ms │            24.07 / 24.18 ±0.18 / 24.55 ms │     no change │
│ QQuery 53 │        29.12 / 29.40 ±0.14 / 29.53 ms │            29.45 / 30.37 ±1.12 / 32.50 ms │     no change │
│ QQuery 54 │        53.19 / 55.53 ±4.15 / 63.82 ms │            53.51 / 55.61 ±3.13 / 61.78 ms │     no change │
│ QQuery 55 │        23.74 / 23.92 ±0.12 / 24.04 ms │            23.94 / 24.08 ±0.16 / 24.38 ms │     no change │
│ QQuery 56 │        37.52 / 37.98 ±0.33 / 38.40 ms │            38.03 / 38.43 ±0.24 / 38.72 ms │     no change │
│ QQuery 57 │     150.65 / 152.89 ±3.34 / 159.52 ms │         151.72 / 154.51 ±2.39 / 158.60 ms │     no change │
│ QQuery 58 │     102.06 / 103.39 ±1.10 / 105.04 ms │         102.73 / 103.60 ±0.58 / 104.40 ms │     no change │
│ QQuery 59 │     106.60 / 107.24 ±0.91 / 109.03 ms │         107.67 / 108.99 ±1.82 / 112.51 ms │     no change │
│ QQuery 60 │        38.14 / 38.89 ±0.48 / 39.55 ms │            37.81 / 38.58 ±0.68 / 39.50 ms │     no change │
│ QQuery 61 │           8.86 / 9.09 ±0.21 / 9.43 ms │               8.95 / 9.13 ±0.20 / 9.50 ms │     no change │
│ QQuery 62 │        37.57 / 37.87 ±0.17 / 38.05 ms │            37.50 / 37.86 ±0.22 / 38.12 ms │     no change │
│ QQuery 63 │        29.43 / 29.58 ±0.11 / 29.69 ms │            29.44 / 29.97 ±0.79 / 31.55 ms │     no change │
│ QQuery 64 │     351.89 / 355.88 ±3.40 / 360.94 ms │         355.07 / 360.02 ±5.91 / 370.29 ms │     no change │
│ QQuery 65 │        75.38 / 76.17 ±0.99 / 78.08 ms │            75.13 / 78.92 ±4.16 / 85.09 ms │     no change │
│ QQuery 66 │        71.15 / 73.07 ±3.02 / 79.03 ms │            71.39 / 72.22 ±0.53 / 72.97 ms │     no change │
│ QQuery 67 │     220.64 / 225.10 ±2.78 / 228.13 ms │         221.75 / 226.19 ±4.29 / 233.36 ms │     no change │
│ QQuery 68 │           6.42 / 6.55 ±0.19 / 6.93 ms │               6.46 / 6.62 ±0.18 / 6.95 ms │     no change │
│ QQuery 69 │        53.35 / 53.73 ±0.21 / 53.99 ms │            53.43 / 54.01 ±0.38 / 54.51 ms │     no change │
│ QQuery 70 │       95.48 / 98.41 ±3.06 / 103.40 ms │          96.59 / 100.30 ±4.42 / 108.71 ms │     no change │
│ QQuery 71 │        34.86 / 35.27 ±0.34 / 35.85 ms │            34.83 / 35.50 ±0.39 / 36.05 ms │     no change │
│ QQuery 72 │ 1571.87 / 1592.80 ±20.38 / 1619.19 ms │     1564.71 / 1609.50 ±44.98 / 1689.14 ms │     no change │
│ QQuery 73 │           5.17 / 5.31 ±0.25 / 5.79 ms │               5.25 / 5.47 ±0.25 / 5.97 ms │     no change │
│ QQuery 74 │     242.94 / 246.01 ±3.75 / 253.19 ms │         239.77 / 245.90 ±4.33 / 251.46 ms │     no change │
│ QQuery 75 │     137.27 / 138.27 ±0.78 / 139.16 ms │         137.78 / 139.59 ±1.26 / 141.34 ms │     no change │
│ QQuery 76 │        34.37 / 34.73 ±0.30 / 35.27 ms │            34.39 / 34.59 ±0.12 / 34.72 ms │     no change │
│ QQuery 77 │        59.49 / 59.91 ±0.34 / 60.31 ms │            59.75 / 60.08 ±0.25 / 60.43 ms │     no change │
│ QQuery 78 │    257.70 / 273.71 ±14.61 / 296.51 ms │        260.85 / 271.77 ±16.42 / 304.42 ms │     no change │
│ QQuery 79 │        66.88 / 67.29 ±0.27 / 67.72 ms │            67.05 / 67.56 ±0.36 / 67.92 ms │     no change │
│ QQuery 80 │        92.34 / 93.62 ±1.20 / 95.33 ms │            93.46 / 95.83 ±2.03 / 99.41 ms │     no change │
│ QQuery 81 │        28.36 / 31.02 ±2.95 / 36.23 ms │            28.17 / 28.32 ±0.18 / 28.63 ms │ +1.10x faster │
│ QQuery 82 │        17.44 / 17.64 ±0.23 / 18.06 ms │            17.04 / 17.47 ±0.54 / 18.51 ms │     no change │
│ QQuery 83 │        32.03 / 32.36 ±0.23 / 32.69 ms │            31.88 / 32.15 ±0.16 / 32.34 ms │     no change │
│ QQuery 84 │        30.45 / 30.80 ±0.25 / 31.13 ms │            30.47 / 30.64 ±0.15 / 30.84 ms │     no change │
│ QQuery 85 │        94.67 / 95.16 ±0.32 / 95.49 ms │           95.17 / 98.76 ±3.26 / 104.41 ms │     no change │
│ QQuery 86 │        24.01 / 25.65 ±1.60 / 28.57 ms │            23.92 / 24.58 ±0.50 / 25.24 ms │     no change │
│ QQuery 87 │        64.51 / 65.05 ±0.47 / 65.73 ms │            64.35 / 64.71 ±0.43 / 65.55 ms │     no change │
│ QQuery 88 │        59.06 / 59.93 ±0.44 / 60.24 ms │            60.56 / 61.00 ±0.42 / 61.58 ms │     no change │
│ QQuery 89 │        35.42 / 35.91 ±0.28 / 36.24 ms │            36.45 / 38.42 ±2.68 / 43.73 ms │  1.07x slower │
│ QQuery 90 │        16.01 / 17.49 ±2.54 / 22.56 ms │            16.41 / 16.57 ±0.18 / 16.86 ms │ +1.06x faster │
│ QQuery 91 │        39.58 / 40.68 ±1.43 / 43.37 ms │            39.96 / 40.58 ±0.52 / 41.54 ms │     no change │
│ QQuery 92 │        28.84 / 29.48 ±0.61 / 30.61 ms │            28.94 / 29.32 ±0.39 / 30.00 ms │     no change │
│ QQuery 93 │        48.70 / 49.91 ±0.79 / 51.02 ms │            49.00 / 49.99 ±0.78 / 50.85 ms │     no change │
│ QQuery 94 │        37.73 / 38.11 ±0.31 / 38.66 ms │            38.29 / 39.75 ±1.51 / 42.61 ms │     no change │
│ QQuery 95 │        78.34 / 79.95 ±1.32 / 81.31 ms │            80.54 / 83.24 ±2.30 / 87.31 ms │     no change │
│ QQuery 96 │        23.44 / 23.70 ±0.22 / 24.04 ms │            23.72 / 23.86 ±0.18 / 24.16 ms │     no change │
│ QQuery 97 │        50.08 / 50.81 ±1.31 / 53.44 ms │            50.58 / 51.19 ±0.76 / 52.67 ms │     no change │
│ QQuery 98 │        37.34 / 38.02 ±0.59 / 39.02 ms │            38.60 / 39.35 ±0.75 / 40.46 ms │     no change │
│ QQuery 99 │        56.13 / 58.34 ±2.66 / 62.93 ms │            56.96 / 58.52 ±2.45 / 63.37 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9045.99ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 9107.55ms │
│ Average Time (HEAD)                                      │   91.37ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   92.00ms │
│ Queries Faster                                           │         2 │
│ Queries Slower                                           │         1 │
│ Queries with No Change                                   │        96 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.2 GiB
Avg memory 1.4 GiB
CPU user 184.4s
CPU sys 5.5s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.0 GiB
Avg memory 1.3 GiB
CPU user 186.7s
CPU sys 5.5s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.17 ms │                                   1.21 ms │     no change │
│ QQuery 1  │   11.75 ms │                                  11.77 ms │     no change │
│ QQuery 2  │   34.55 ms │                                  34.80 ms │     no change │
│ QQuery 3  │   30.79 ms │                                  30.68 ms │     no change │
│ QQuery 4  │  232.24 ms │                                 236.36 ms │     no change │
│ QQuery 5  │  269.78 ms │                                 271.83 ms │     no change │
│ QQuery 6  │    1.22 ms │                                   1.18 ms │     no change │
│ QQuery 7  │   13.04 ms │                                  12.91 ms │     no change │
│ QQuery 8  │  325.42 ms │                                 323.18 ms │     no change │
│ QQuery 9  │  459.25 ms │                                 453.71 ms │     no change │
│ QQuery 10 │   64.30 ms │                                  63.92 ms │     no change │
│ QQuery 11 │   75.52 ms │                                  75.52 ms │     no change │
│ QQuery 12 │  258.46 ms │                                 260.63 ms │     no change │
│ QQuery 13 │  364.41 ms │                                 360.59 ms │     no change │
│ QQuery 14 │  279.59 ms │                                 278.53 ms │     no change │
│ QQuery 15 │  275.22 ms │                                 283.30 ms │     no change │
│ QQuery 16 │  612.75 ms │                                 615.12 ms │     no change │
│ QQuery 17 │  617.65 ms │                                 628.30 ms │     no change │
│ QQuery 18 │ 1213.65 ms │                                1205.21 ms │     no change │
│ QQuery 19 │   27.75 ms │                                  27.77 ms │     no change │
│ QQuery 20 │  514.52 ms │                                 519.75 ms │     no change │
│ QQuery 21 │  506.72 ms │                                 512.87 ms │     no change │
│ QQuery 22 │  976.21 ms │                                 973.39 ms │     no change │
│ QQuery 23 │ 3036.53 ms │                                3064.95 ms │     no change │
│ QQuery 24 │   40.26 ms │                                  40.90 ms │     no change │
│ QQuery 25 │  105.86 ms │                                 105.32 ms │     no change │
│ QQuery 26 │   40.84 ms │                                  40.94 ms │     no change │
│ QQuery 27 │  510.67 ms │                                 514.17 ms │     no change │
│ QQuery 28 │ 2905.97 ms │                                2847.86 ms │     no change │
│ QQuery 29 │   37.78 ms │                                  37.93 ms │     no change │
│ QQuery 30 │  299.83 ms │                                 304.67 ms │     no change │
│ QQuery 31 │  277.11 ms │                                 282.37 ms │     no change │
│ QQuery 32 │  919.26 ms │                                 938.97 ms │     no change │
│ QQuery 33 │ 1447.07 ms │                                1430.88 ms │     no change │
│ QQuery 34 │ 1443.58 ms │                                1476.65 ms │     no change │
│ QQuery 35 │  284.69 ms │                                 284.32 ms │     no change │
│ QQuery 36 │   66.37 ms │                                  65.61 ms │     no change │
│ QQuery 37 │   34.20 ms │                                  34.04 ms │     no change │
│ QQuery 38 │   39.34 ms │                                  38.65 ms │     no change │
│ QQuery 39 │  148.14 ms │                                 130.70 ms │ +1.13x faster │
│ QQuery 40 │   13.47 ms │                                  12.88 ms │     no change │
│ QQuery 41 │   12.57 ms │                                  12.37 ms │     no change │
│ QQuery 42 │   11.72 ms │                                  11.90 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 18841.18ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 18858.60ms │
│ Average Time (HEAD)                                      │   438.17ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   438.57ms │
│ Queries Faster                                           │          1 │
│ Queries Slower                                           │          0 │
│ Queries with No Change                                   │         42 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │           1.17 / 1.78 ±1.12 / 4.00 ms │               1.21 / 1.80 ±1.08 / 3.96 ms │     no change │
│ QQuery 1  │        11.75 / 11.86 ±0.10 / 12.01 ms │            11.77 / 12.16 ±0.53 / 13.20 ms │     no change │
│ QQuery 2  │        34.55 / 35.06 ±0.57 / 36.16 ms │            34.80 / 35.07 ±0.19 / 35.35 ms │     no change │
│ QQuery 3  │        30.79 / 32.14 ±1.73 / 35.31 ms │            30.68 / 30.97 ±0.29 / 31.51 ms │     no change │
│ QQuery 4  │     232.24 / 236.28 ±3.07 / 240.34 ms │         236.36 / 238.20 ±1.61 / 240.44 ms │     no change │
│ QQuery 5  │     269.78 / 274.34 ±3.76 / 279.64 ms │         271.83 / 274.57 ±1.81 / 277.41 ms │     no change │
│ QQuery 6  │           1.22 / 1.37 ±0.24 / 1.84 ms │               1.18 / 1.35 ±0.26 / 1.86 ms │     no change │
│ QQuery 7  │        13.04 / 13.13 ±0.07 / 13.23 ms │            12.91 / 13.13 ±0.13 / 13.28 ms │     no change │
│ QQuery 8  │     325.42 / 326.57 ±1.30 / 328.98 ms │         323.18 / 327.96 ±4.80 / 333.97 ms │     no change │
│ QQuery 9  │     459.25 / 468.76 ±6.22 / 477.45 ms │        453.71 / 469.70 ±12.38 / 491.11 ms │     no change │
│ QQuery 10 │        64.30 / 65.12 ±1.00 / 67.03 ms │            63.92 / 67.68 ±4.57 / 76.65 ms │     no change │
│ QQuery 11 │        75.52 / 76.04 ±0.56 / 77.13 ms │            75.52 / 77.94 ±3.52 / 84.94 ms │     no change │
│ QQuery 12 │     258.46 / 263.65 ±4.96 / 272.78 ms │         260.63 / 265.98 ±3.54 / 270.33 ms │     no change │
│ QQuery 13 │     364.41 / 371.83 ±6.56 / 382.46 ms │        360.59 / 372.20 ±10.75 / 391.96 ms │     no change │
│ QQuery 14 │     279.59 / 284.34 ±6.54 / 297.25 ms │         278.53 / 281.99 ±1.93 / 283.64 ms │     no change │
│ QQuery 15 │    275.22 / 289.55 ±11.95 / 307.23 ms │         283.30 / 290.76 ±5.43 / 300.25 ms │     no change │
│ QQuery 16 │     612.75 / 620.55 ±4.75 / 627.10 ms │        615.12 / 634.44 ±23.39 / 680.00 ms │     no change │
│ QQuery 17 │     617.65 / 623.74 ±6.11 / 634.18 ms │        628.30 / 636.14 ±11.58 / 659.18 ms │     no change │
│ QQuery 18 │ 1213.65 / 1235.18 ±11.32 / 1245.03 ms │     1205.21 / 1223.17 ±14.66 / 1236.90 ms │     no change │
│ QQuery 19 │       27.75 / 38.16 ±15.93 / 69.73 ms │            27.77 / 38.35 ±9.20 / 51.80 ms │     no change │
│ QQuery 20 │    514.52 / 522.65 ±10.70 / 543.61 ms │         519.75 / 528.44 ±7.61 / 539.76 ms │     no change │
│ QQuery 21 │     506.72 / 515.13 ±5.65 / 521.89 ms │         512.87 / 515.73 ±3.87 / 523.32 ms │     no change │
│ QQuery 22 │     976.21 / 980.65 ±3.93 / 986.87 ms │       973.39 / 986.22 ±16.06 / 1016.70 ms │     no change │
│ QQuery 23 │ 3036.53 / 3069.31 ±18.17 / 3086.62 ms │     3064.95 / 3089.82 ±24.97 / 3121.41 ms │     no change │
│ QQuery 24 │        40.26 / 40.49 ±0.22 / 40.82 ms │           40.90 / 50.95 ±12.32 / 72.57 ms │  1.26x slower │
│ QQuery 25 │     105.86 / 110.37 ±4.26 / 117.38 ms │         105.32 / 108.81 ±2.98 / 113.90 ms │     no change │
│ QQuery 26 │        40.84 / 41.47 ±0.53 / 42.43 ms │            40.94 / 43.74 ±4.08 / 51.84 ms │  1.05x slower │
│ QQuery 27 │     510.67 / 514.20 ±4.43 / 522.89 ms │         514.17 / 517.45 ±2.94 / 522.97 ms │     no change │
│ QQuery 28 │ 2905.97 / 2919.20 ±14.54 / 2946.76 ms │     2847.86 / 2873.26 ±16.38 / 2894.27 ms │     no change │
│ QQuery 29 │        37.78 / 38.10 ±0.28 / 38.62 ms │            37.93 / 45.31 ±6.11 / 54.43 ms │  1.19x slower │
│ QQuery 30 │     299.83 / 308.88 ±8.91 / 325.83 ms │         304.67 / 309.57 ±4.34 / 317.05 ms │     no change │
│ QQuery 31 │     277.11 / 287.14 ±5.55 / 294.15 ms │         282.37 / 285.10 ±1.56 / 287.22 ms │     no change │
│ QQuery 32 │    919.26 / 944.19 ±17.62 / 972.53 ms │       938.97 / 974.05 ±32.42 / 1035.26 ms │     no change │
│ QQuery 33 │ 1447.07 / 1488.66 ±34.48 / 1551.12 ms │     1430.88 / 1461.57 ±20.60 / 1488.73 ms │     no change │
│ QQuery 34 │ 1443.58 / 1483.19 ±31.74 / 1527.23 ms │     1476.65 / 1498.48 ±17.52 / 1527.18 ms │     no change │
│ QQuery 35 │    284.69 / 312.76 ±35.43 / 377.67 ms │        284.32 / 327.53 ±57.38 / 436.80 ms │     no change │
│ QQuery 36 │        66.37 / 74.91 ±8.29 / 90.40 ms │            65.61 / 69.79 ±4.72 / 77.45 ms │ +1.07x faster │
│ QQuery 37 │        34.20 / 37.81 ±5.12 / 47.92 ms │            34.04 / 34.44 ±0.41 / 35.15 ms │ +1.10x faster │
│ QQuery 38 │        39.34 / 42.02 ±1.59 / 43.82 ms │            38.65 / 39.99 ±1.67 / 42.78 ms │     no change │
│ QQuery 39 │     148.14 / 156.31 ±4.66 / 162.17 ms │        130.70 / 145.46 ±12.14 / 159.83 ms │ +1.07x faster │
│ QQuery 40 │        13.47 / 14.06 ±0.51 / 14.87 ms │            12.88 / 13.18 ±0.19 / 13.46 ms │ +1.07x faster │
│ QQuery 41 │        12.57 / 12.85 ±0.33 / 13.46 ms │            12.37 / 12.48 ±0.09 / 12.60 ms │     no change │
│ QQuery 42 │        11.72 / 13.70 ±3.05 / 19.75 ms │            11.90 / 13.10 ±2.13 / 17.35 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 19197.52ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 19238.03ms │
│ Average Time (HEAD)                                      │   446.45ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   447.40ms │
│ Queries Faster                                           │          4 │
│ Queries Slower                                           │          3 │
│ Queries with No Change                                   │         36 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 100.0s
Peak memory 11.5 GiB
Avg memory 4.2 GiB
CPU user 984.0s
CPU sys 70.8s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 100.0s
Peak memory 11.7 GiB
Avg memory 4.5 GiB
CPU user 981.2s
CPU sys 71.0s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.17 ms │                                   1.14 ms │     no change │
│ QQuery 1  │   11.72 ms │                                  11.55 ms │     no change │
│ QQuery 2  │   34.76 ms │                                  34.40 ms │     no change │
│ QQuery 3  │   30.58 ms │                                  30.63 ms │     no change │
│ QQuery 4  │  241.80 ms │                                 245.23 ms │     no change │
│ QQuery 5  │  838.84 ms │                                 757.37 ms │ +1.11x faster │
│ QQuery 6  │    1.18 ms │                                   1.16 ms │     no change │
│ QQuery 7  │   12.98 ms │                                  12.99 ms │     no change │
│ QQuery 8  │  346.81 ms │                                 347.93 ms │     no change │
│ QQuery 9  │  479.76 ms │                                 495.70 ms │     no change │
│ QQuery 10 │  202.05 ms │                                 201.46 ms │     no change │
│ QQuery 11 │  213.14 ms │                                 212.90 ms │     no change │
│ QQuery 12 │  833.96 ms │                                 788.19 ms │ +1.06x faster │
│ QQuery 13 │ 1119.10 ms │                                1088.21 ms │     no change │
│ QQuery 14 │  860.35 ms │                                 888.48 ms │     no change │
│ QQuery 15 │  306.17 ms │                                 293.40 ms │     no change │
│ QQuery 16 │ 1335.10 ms │                                1352.74 ms │     no change │
│ QQuery 17 │ 1283.21 ms │                                1303.19 ms │     no change │
│ QQuery 18 │ 2042.59 ms │                                2098.54 ms │     no change │
│ QQuery 19 │   28.07 ms │                                  27.86 ms │     no change │
│ QQuery 20 │ 1061.34 ms │                                1065.80 ms │     no change │
│ QQuery 21 │ 1387.75 ms │                                1378.84 ms │     no change │
│ QQuery 22 │ 3295.60 ms │                                3267.97 ms │     no change │
│ QQuery 23 │ 9253.37 ms │                                9290.22 ms │     no change │
│ QQuery 24 │   95.46 ms │                                  96.88 ms │     no change │
│ QQuery 25 │  358.39 ms │                                 354.99 ms │     no change │
│ QQuery 26 │   99.53 ms │                                  98.44 ms │     no change │
│ QQuery 27 │ 1181.04 ms │                                1159.91 ms │     no change │
│ QQuery 28 │ 2570.54 ms │                                2509.17 ms │     no change │
│ QQuery 29 │   38.19 ms │                                  37.59 ms │     no change │
│ QQuery 30 │  550.71 ms │                                 553.88 ms │     no change │
│ QQuery 31 │  535.48 ms │                                 551.07 ms │     no change │
│ QQuery 32 │ 1058.15 ms │                                1050.40 ms │     no change │
│ QQuery 33 │ 3827.75 ms │                                3652.59 ms │     no change │
│ QQuery 34 │ 3888.92 ms │                                3606.83 ms │ +1.08x faster │
│ QQuery 35 │  312.25 ms │                                 318.00 ms │     no change │
│ QQuery 36 │  136.39 ms │                                 128.55 ms │ +1.06x faster │
│ QQuery 37 │   94.83 ms │                                  94.21 ms │     no change │
│ QQuery 38 │   82.57 ms │                                  81.48 ms │     no change │
│ QQuery 39 │  321.63 ms │                                 297.62 ms │ +1.08x faster │
│ QQuery 40 │   13.89 ms │                                  13.68 ms │     no change │
│ QQuery 41 │   13.17 ms │                                  13.07 ms │     no change │
│ QQuery 42 │   12.66 ms │                                  12.59 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 40412.96ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 39826.84ms │
│ Average Time (HEAD)                                      │   939.84ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   926.21ms │
│ Queries Faster                                           │          5 │
│ Queries Slower                                           │          0 │
│ Queries with No Change                                   │         38 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                   HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │            1.17 / 1.82 ±1.16 / 4.12 ms │               1.14 / 1.79 ±1.16 / 4.10 ms │     no change │
│ QQuery 1  │         11.72 / 11.98 ±0.17 / 12.18 ms │            11.55 / 11.77 ±0.12 / 11.89 ms │     no change │
│ QQuery 2  │         34.76 / 35.55 ±0.93 / 37.35 ms │            34.40 / 34.87 ±0.44 / 35.52 ms │     no change │
│ QQuery 3  │         30.58 / 31.32 ±0.78 / 32.70 ms │            30.63 / 30.96 ±0.54 / 32.03 ms │     no change │
│ QQuery 4  │      241.80 / 246.92 ±3.32 / 250.22 ms │         245.23 / 248.17 ±2.09 / 250.12 ms │     no change │
│ QQuery 5  │      838.84 / 844.40 ±4.78 / 851.46 ms │        757.37 / 781.58 ±14.58 / 803.01 ms │ +1.08x faster │
│ QQuery 6  │            1.18 / 1.33 ±0.24 / 1.80 ms │               1.16 / 1.32 ±0.25 / 1.80 ms │     no change │
│ QQuery 7  │         12.98 / 13.24 ±0.14 / 13.36 ms │            12.99 / 13.08 ±0.07 / 13.16 ms │     no change │
│ QQuery 8  │      346.81 / 349.50 ±2.23 / 352.76 ms │         347.93 / 352.23 ±3.34 / 357.60 ms │     no change │
│ QQuery 9  │     479.76 / 495.88 ±11.00 / 513.95 ms │         495.70 / 501.44 ±5.17 / 509.83 ms │     no change │
│ QQuery 10 │      202.05 / 203.19 ±1.31 / 205.22 ms │         201.46 / 203.41 ±1.64 / 205.26 ms │     no change │
│ QQuery 11 │      213.14 / 214.18 ±0.60 / 214.93 ms │         212.90 / 215.18 ±3.18 / 221.39 ms │     no change │
│ QQuery 12 │      833.96 / 843.44 ±6.95 / 855.45 ms │        788.19 / 813.03 ±12.79 / 822.51 ms │     no change │
│ QQuery 13 │  1119.10 / 1143.32 ±14.89 / 1162.13 ms │     1088.21 / 1133.60 ±41.38 / 1211.69 ms │     no change │
│ QQuery 14 │     860.35 / 894.29 ±18.10 / 913.71 ms │        888.48 / 904.95 ±16.86 / 936.14 ms │     no change │
│ QQuery 15 │      306.17 / 309.17 ±3.71 / 316.07 ms │        293.40 / 312.63 ±11.74 / 329.52 ms │     no change │
│ QQuery 16 │  1335.10 / 1347.51 ±12.47 / 1367.98 ms │      1352.74 / 1367.15 ±8.90 / 1378.24 ms │     no change │
│ QQuery 17 │  1283.21 / 1312.03 ±23.64 / 1346.49 ms │     1303.19 / 1331.24 ±16.26 / 1349.60 ms │     no change │
│ QQuery 18 │  2042.59 / 2093.97 ±26.78 / 2120.17 ms │     2098.54 / 2123.88 ±15.58 / 2141.66 ms │     no change │
│ QQuery 19 │         28.07 / 29.20 ±1.18 / 30.85 ms │            27.86 / 28.53 ±0.59 / 29.54 ms │     no change │
│ QQuery 20 │   1061.34 / 1074.61 ±7.88 / 1082.46 ms │      1065.80 / 1074.19 ±7.38 / 1086.73 ms │     no change │
│ QQuery 21 │  1387.75 / 1399.13 ±11.24 / 1420.03 ms │      1378.84 / 1385.51 ±6.00 / 1396.32 ms │     no change │
│ QQuery 22 │  3295.60 / 3338.04 ±37.47 / 3393.63 ms │     3267.97 / 3320.31 ±61.70 / 3437.48 ms │     no change │
│ QQuery 23 │ 9253.37 / 9518.24 ±160.03 / 9737.33 ms │    9290.22 / 9514.12 ±138.16 / 9723.20 ms │     no change │
│ QQuery 24 │       95.46 / 101.43 ±7.57 / 116.35 ms │         96.88 / 116.91 ±36.71 / 190.27 ms │  1.15x slower │
│ QQuery 25 │      358.39 / 363.03 ±5.79 / 374.13 ms │        354.99 / 377.10 ±41.14 / 459.36 ms │     no change │
│ QQuery 26 │       99.53 / 100.66 ±0.92 / 102.22 ms │           98.44 / 99.93 ±1.30 / 101.86 ms │     no change │
│ QQuery 27 │  1181.04 / 1194.10 ±14.48 / 1222.35 ms │      1159.91 / 1173.42 ±8.07 / 1181.52 ms │     no change │
│ QQuery 28 │  2570.54 / 2602.35 ±23.99 / 2633.77 ms │     2509.17 / 2538.61 ±15.22 / 2553.02 ms │     no change │
│ QQuery 29 │       38.19 / 51.23 ±25.21 / 101.64 ms │            37.59 / 38.38 ±1.10 / 40.55 ms │ +1.33x faster │
│ QQuery 30 │     550.71 / 573.69 ±25.34 / 618.49 ms │         553.88 / 559.48 ±6.10 / 571.28 ms │     no change │
│ QQuery 31 │     535.48 / 562.29 ±16.10 / 582.88 ms │        551.07 / 568.37 ±10.32 / 580.94 ms │     no change │
│ QQuery 32 │ 1058.15 / 1157.28 ±103.41 / 1349.73 ms │     1050.40 / 1129.83 ±65.24 / 1244.97 ms │     no change │
│ QQuery 33 │  3827.75 / 3899.14 ±57.33 / 3992.84 ms │     3652.59 / 3700.79 ±29.16 / 3732.26 ms │ +1.05x faster │
│ QQuery 34 │  3888.92 / 4014.60 ±82.02 / 4117.36 ms │     3606.83 / 3754.55 ±77.29 / 3831.89 ms │ +1.07x faster │
│ QQuery 35 │      312.25 / 324.00 ±9.13 / 335.22 ms │        318.00 / 339.57 ±22.52 / 381.22 ms │     no change │
│ QQuery 36 │      136.39 / 149.75 ±7.70 / 158.18 ms │         128.55 / 138.80 ±6.00 / 146.33 ms │ +1.08x faster │
│ QQuery 37 │        94.83 / 97.28 ±1.96 / 100.59 ms │          94.21 / 102.70 ±9.13 / 119.89 ms │  1.06x slower │
│ QQuery 38 │         82.57 / 86.54 ±3.02 / 89.69 ms │            81.48 / 86.86 ±5.05 / 93.17 ms │     no change │
│ QQuery 39 │      321.63 / 332.84 ±6.35 / 339.42 ms │        297.62 / 328.18 ±16.06 / 343.00 ms │     no change │
│ QQuery 40 │         13.89 / 16.20 ±2.59 / 21.23 ms │            13.68 / 17.52 ±6.92 / 31.33 ms │  1.08x slower │
│ QQuery 41 │         13.17 / 14.27 ±1.72 / 17.66 ms │            13.07 / 13.39 ±0.28 / 13.87 ms │ +1.07x faster │
│ QQuery 42 │         12.66 / 15.25 ±5.00 / 25.24 ms │            12.59 / 12.96 ±0.35 / 13.61 ms │ +1.18x faster │
└───────────┴────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 41408.18ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 40802.28ms │
│ Average Time (HEAD)                                      │   962.98ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   948.89ms │
│ Queries Faster                                           │          7 │
│ Queries Slower                                           │          3 │
│ Queries with No Change                                   │         33 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 210.1s
Peak memory 13.1 GiB
Avg memory 5.7 GiB
CPU user 2005.5s
CPU sys 141.5s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 205.1s
Peak memory 12.7 GiB
Avg memory 5.3 GiB
CPU user 1965.3s
CPU sys 139.1s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (0d75b53) to 14b3edd (merge-base) diff

Run configuration
run benchmark dictionary_group_values
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

group                                                                             HEAD                                    rich-t-kid_single-dictionary-group-values
-----                                                                             ----                                    -----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       2.06    479.0±3.05µs 130.5 MElem/sec    1.00    232.2±6.76µs 269.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         2.13    414.3±2.33µs 150.9 MElem/sec    1.00    194.7±6.97µs 321.0 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        2.14    440.1±3.92µs 142.0 MElem/sec    1.00    206.1±7.04µs 303.3 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.27      6.0±0.02ms 10.4 MElem/sec     1.00      4.7±0.02ms 13.3 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         2.19    431.7±2.00µs 144.8 MElem/sec    1.00    197.4±7.22µs 316.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.74    111.6±0.69µs 70.0 MElem/sec     1.00     64.1±0.51µs 121.8 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          2.05     55.7±0.55µs 140.4 MElem/sec    1.00     27.1±9.44µs 288.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.92     72.7±0.48µs 107.5 MElem/sec    1.00     37.8±2.53µs 206.9 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          2.12     61.3±0.74µs 127.5 MElem/sec    1.00     28.9±0.63µs 270.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.49    523.0±2.83µs 14.9 MElem/sec     1.00    350.8±0.91µs 22.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.84      2.2±0.01ms 113.7 MElem/sec    1.00   1193.2±4.87µs 209.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.90      2.0±0.01ms 122.8 MElem/sec    1.00   1072.4±5.65µs 233.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.93      2.1±0.01ms 117.9 MElem/sec    1.00   1101.1±4.84µs 227.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.00     19.1±0.10ms 13.1 MElem/sec     1.10     21.0±0.10ms 11.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.96      2.1±0.00ms 118.6 MElem/sec    1.00   1078.0±4.42µs 231.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.49    378.7±1.55µs 82.5 MElem/sec     1.00    253.8±1.13µs 123.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.80    248.4±1.23µs 125.8 MElem/sec    1.00    137.6±0.72µs 227.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.77    287.0±1.66µs 108.9 MElem/sec    1.00    162.5±0.57µs 192.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.84    262.1±1.30µs 119.2 MElem/sec    1.00    142.6±0.65µs 219.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.19   1289.5±9.43µs 24.2 MElem/sec     1.00   1081.3±5.25µs 28.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       4.17   1341.6±4.71µs 46.6 MElem/sec     1.00    321.4±1.60µs 194.4 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         4.55   1245.1±3.12µs 50.2 MElem/sec     1.00    273.9±1.22µs 228.2 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        4.32   1245.8±3.96µs 50.2 MElem/sec     1.00    288.2±1.25µs 216.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.34      7.4±0.02ms  8.4 MElem/sec     1.00      5.5±0.02ms 11.3 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         4.44   1231.3±5.96µs 50.8 MElem/sec     1.00    277.5±1.77µs 225.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.99     18.7±0.08ms 13.3 MElem/sec     1.00      9.4±0.03ms 26.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     2.48      7.9±0.01ms 31.6 MElem/sec     1.00      3.2±0.02ms 78.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    2.15     64.6±0.85ms  3.9 MElem/sec     1.00     30.1±0.97ms  8.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.94     13.7±0.03ms 18.2 MElem/sec     1.00      7.1±0.04ms 35.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      2.27      2.8±0.01ms 88.8 MElem/sec     1.00  1241.5±14.97µs 201.4 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.77      2.3±0.01ms 106.9 MElem/sec    1.00   1322.3±7.23µs 189.1 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.85      2.0±0.01ms 124.7 MElem/sec    1.00   1082.3±5.15µs 231.0 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.86      2.1±0.00ms 117.0 MElem/sec    1.00   1147.0±4.50µs 218.0 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.12     26.0±0.16ms  9.6 MElem/sec     1.00     23.2±0.05ms 10.8 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.90      2.1±0.00ms 120.1 MElem/sec    1.00   1097.0±4.72µs 227.9 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric Value
Wall time 610.1s
Peak memory 886.5 MiB
Avg memory 66.6 MiB
CPU user 402.7s
CPU sys 23.2s
Peak spill 0 B

dictionary_group_values — branch

Metric Value
Wall time 630.2s
Peak memory 516.0 MiB
Avg memory 162.9 MiB
CPU user 442.7s
CPU sys 5.4s
Peak spill 0 B

File an issue against this benchmark runner

@Rich-T-kid
Rich-T-kid force-pushed the rich-t-kid/single-dictionary-group-values branch from 0d75b53 to 8c55192 Compare November 9, 2026 23:22
@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark dictionary_group_values

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6754176866-4722-2crgc 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (8c55192) to 14b3edd (merge-base) diff

Run configuration
run benchmark dictionary_group_values

Results will be posted here when complete


File an issue against this benchmark runner

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark tpch tpcds clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6754328293-4724-vqn8n 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (8c55192) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6754328293-4723-v5l22 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (8c55192) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6754328293-4725-7mfgz 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (8c55192) to 14b3edd (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@Rich-T-kid
Rich-T-kid marked this pull request as ready for review November 9, 2026 23:36
@Rich-T-kid
Rich-T-kid marked this pull request as draft November 9, 2026 23:36
@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (8c55192) to 14b3edd (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 42.03 ms │                                  40.47 ms │     no change │
│ QQuery 2  │ 20.66 ms │                                  19.94 ms │     no change │
│ QQuery 3  │ 30.80 ms │                                  30.65 ms │     no change │
│ QQuery 4  │ 18.74 ms │                                  19.63 ms │     no change │
│ QQuery 5  │ 39.08 ms │                                  38.20 ms │     no change │
│ QQuery 6  │ 14.75 ms │                                  14.91 ms │     no change │
│ QQuery 7  │ 44.44 ms │                                  46.16 ms │     no change │
│ QQuery 8  │ 43.05 ms │                                  41.63 ms │     no change │
│ QQuery 9  │ 52.71 ms │                                  52.67 ms │     no change │
│ QQuery 10 │ 62.27 ms │                                  61.76 ms │     no change │
│ QQuery 11 │ 15.22 ms │                                  15.60 ms │     no change │
│ QQuery 12 │ 21.73 ms │                                  22.63 ms │     no change │
│ QQuery 13 │ 53.52 ms │                                  54.49 ms │     no change │
│ QQuery 14 │ 25.24 ms │                                  26.29 ms │     no change │
│ QQuery 15 │ 32.37 ms │                                  33.20 ms │     no change │
│ QQuery 16 │ 14.13 ms │                                  14.27 ms │     no change │
│ QQuery 17 │ 76.97 ms │                                  77.74 ms │     no change │
│ QQuery 18 │ 77.83 ms │                                  70.88 ms │ +1.10x faster │
│ QQuery 19 │ 31.46 ms │                                  30.91 ms │     no change │
│ QQuery 20 │ 37.32 ms │                                  37.43 ms │     no change │
│ QQuery 21 │ 60.83 ms │                                  59.39 ms │     no change │
│ QQuery 22 │ 17.09 ms │                                  16.98 ms │     no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 832.25ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 825.82ms │
│ Average Time (HEAD)                                      │  37.83ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  37.54ms │
│ Queries Faster                                           │        1 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       21 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 42.03 / 42.97 ±1.26 / 45.33 ms │            40.47 / 41.35 ±1.21 / 43.72 ms │     no change │
│ QQuery 2  │ 20.66 / 20.90 ±0.32 / 21.50 ms │            19.94 / 20.19 ±0.22 / 20.47 ms │     no change │
│ QQuery 3  │ 30.80 / 32.41 ±2.87 / 38.13 ms │            30.65 / 31.21 ±0.49 / 32.09 ms │     no change │
│ QQuery 4  │ 18.74 / 19.48 ±0.75 / 20.45 ms │            19.63 / 21.63 ±2.52 / 26.54 ms │  1.11x slower │
│ QQuery 5  │ 39.08 / 39.89 ±0.60 / 40.91 ms │            38.20 / 39.14 ±0.91 / 40.63 ms │     no change │
│ QQuery 6  │ 14.75 / 15.09 ±0.52 / 16.12 ms │            14.91 / 15.21 ±0.43 / 16.05 ms │     no change │
│ QQuery 7  │ 44.44 / 47.94 ±2.97 / 53.37 ms │            46.16 / 48.11 ±2.32 / 52.65 ms │     no change │
│ QQuery 8  │ 43.05 / 43.82 ±1.22 / 46.24 ms │            41.63 / 42.08 ±0.28 / 42.33 ms │     no change │
│ QQuery 9  │ 52.71 / 53.66 ±0.67 / 54.73 ms │            52.67 / 53.54 ±0.71 / 54.48 ms │     no change │
│ QQuery 10 │ 62.27 / 65.17 ±2.63 / 69.13 ms │            61.76 / 63.73 ±1.65 / 66.63 ms │     no change │
│ QQuery 11 │ 15.22 / 15.37 ±0.16 / 15.58 ms │            15.60 / 15.94 ±0.47 / 16.86 ms │     no change │
│ QQuery 12 │ 21.73 / 23.08 ±1.80 / 26.63 ms │            22.63 / 24.55 ±3.20 / 30.94 ms │  1.06x slower │
│ QQuery 13 │ 53.52 / 54.41 ±0.62 / 55.13 ms │            54.49 / 54.95 ±0.43 / 55.69 ms │     no change │
│ QQuery 14 │ 25.24 / 25.55 ±0.22 / 25.81 ms │            26.29 / 26.43 ±0.21 / 26.85 ms │     no change │
│ QQuery 15 │ 32.37 / 32.96 ±0.79 / 34.39 ms │            33.20 / 34.14 ±0.90 / 35.35 ms │     no change │
│ QQuery 16 │ 14.13 / 14.19 ±0.05 / 14.27 ms │            14.27 / 14.40 ±0.07 / 14.50 ms │     no change │
│ QQuery 17 │ 76.97 / 80.36 ±3.57 / 86.86 ms │            77.74 / 84.34 ±5.91 / 93.39 ms │     no change │
│ QQuery 18 │ 77.83 / 79.26 ±0.95 / 80.54 ms │            70.88 / 74.78 ±3.22 / 78.23 ms │ +1.06x faster │
│ QQuery 19 │ 31.46 / 32.00 ±0.55 / 33.05 ms │            30.91 / 31.75 ±1.03 / 33.76 ms │     no change │
│ QQuery 20 │ 37.32 / 37.67 ±0.23 / 37.90 ms │            37.43 / 38.07 ±0.35 / 38.39 ms │     no change │
│ QQuery 21 │ 60.83 / 62.92 ±1.45 / 64.69 ms │            59.39 / 62.55 ±3.18 / 68.34 ms │     no change │
│ QQuery 22 │ 17.09 / 17.21 ±0.18 / 17.56 ms │            16.98 / 17.29 ±0.19 / 17.57 ms │     no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 856.32ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 855.38ms │
│ Average Time (HEAD)                                      │  38.92ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  38.88ms │
│ Queries Faster                                           │        1 │
│ Queries Slower                                           │        2 │
│ Queries with No Change                                   │       19 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 656.3 MiB
CPU user 21.1s
CPU sys 1.8s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 647.0 MiB
CPU user 21.2s
CPU sys 1.8s
Peak spill 0 B

File an issue against this benchmark runner

@Rich-T-kid
Rich-T-kid force-pushed the rich-t-kid/single-dictionary-group-values branch from f34451c to f6ce571 Compare November 10, 2026 04:04
@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark dictionary_group_values

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6758664863-4737-rm5jm 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark dictionary_group_values

Results will be posted here when complete


File an issue against this benchmark runner

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark tpch tpcds clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6758715157-4740-bfrgt 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6758715157-4738-spblz 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6758715157-4739-7z7wc 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 39.21 ms │                                  39.30 ms │ no change │
│ QQuery 2  │ 18.72 ms │                                  18.59 ms │ no change │
│ QQuery 3  │ 28.04 ms │                                  28.37 ms │ no change │
│ QQuery 4  │ 16.79 ms │                                  16.82 ms │ no change │
│ QQuery 5  │ 35.06 ms │                                  35.14 ms │ no change │
│ QQuery 6  │ 13.66 ms │                                  13.61 ms │ no change │
│ QQuery 7  │ 41.24 ms │                                  41.32 ms │ no change │
│ QQuery 8  │ 40.88 ms │                                  40.57 ms │ no change │
│ QQuery 9  │ 48.66 ms │                                  50.25 ms │ no change │
│ QQuery 10 │ 58.73 ms │                                  58.41 ms │ no change │
│ QQuery 11 │ 13.61 ms │                                  13.62 ms │ no change │
│ QQuery 12 │ 20.79 ms │                                  20.33 ms │ no change │
│ QQuery 13 │ 45.69 ms │                                  45.56 ms │ no change │
│ QQuery 14 │ 23.80 ms │                                  23.78 ms │ no change │
│ QQuery 15 │ 30.82 ms │                                  31.23 ms │ no change │
│ QQuery 16 │ 13.32 ms │                                  13.65 ms │ no change │
│ QQuery 17 │ 67.48 ms │                                  67.94 ms │ no change │
│ QQuery 18 │ 64.30 ms │                                  63.70 ms │ no change │
│ QQuery 19 │ 29.31 ms │                                  28.93 ms │ no change │
│ QQuery 20 │ 33.77 ms │                                  33.69 ms │ no change │
│ QQuery 21 │ 52.99 ms │                                  52.27 ms │ no change │
│ QQuery 22 │ 15.63 ms │                                  15.34 ms │ no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 752.50ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 752.40ms │
│ Average Time (HEAD)                                      │  34.20ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  34.20ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 39.21 / 40.04 ±1.07 / 42.14 ms │            39.30 / 40.29 ±1.25 / 42.60 ms │ no change │
│ QQuery 2  │ 18.72 / 19.22 ±0.57 / 20.28 ms │            18.59 / 18.87 ±0.28 / 19.40 ms │ no change │
│ QQuery 3  │ 28.04 / 28.42 ±0.27 / 28.84 ms │            28.37 / 28.57 ±0.22 / 28.98 ms │ no change │
│ QQuery 4  │ 16.79 / 17.73 ±0.70 / 18.63 ms │            16.82 / 16.88 ±0.04 / 16.92 ms │ no change │
│ QQuery 5  │ 35.06 / 35.46 ±0.40 / 36.20 ms │            35.14 / 35.42 ±0.40 / 36.20 ms │ no change │
│ QQuery 6  │ 13.66 / 13.86 ±0.30 / 14.46 ms │            13.61 / 13.74 ±0.09 / 13.86 ms │ no change │
│ QQuery 7  │ 41.24 / 43.44 ±1.94 / 45.64 ms │            41.32 / 42.69 ±1.12 / 44.07 ms │ no change │
│ QQuery 8  │ 40.88 / 41.99 ±2.01 / 46.01 ms │            40.57 / 41.41 ±1.01 / 43.35 ms │ no change │
│ QQuery 9  │ 48.66 / 50.01 ±1.26 / 52.21 ms │            50.25 / 51.59 ±0.87 / 52.66 ms │ no change │
│ QQuery 10 │ 58.73 / 59.39 ±0.68 / 60.47 ms │            58.41 / 59.69 ±0.83 / 60.66 ms │ no change │
│ QQuery 11 │ 13.61 / 13.86 ±0.20 / 14.20 ms │            13.62 / 13.88 ±0.16 / 14.12 ms │ no change │
│ QQuery 12 │ 20.79 / 20.92 ±0.14 / 21.18 ms │            20.33 / 20.62 ±0.19 / 20.86 ms │ no change │
│ QQuery 13 │ 45.69 / 46.81 ±1.46 / 49.59 ms │            45.56 / 46.49 ±1.25 / 48.93 ms │ no change │
│ QQuery 14 │ 23.80 / 23.91 ±0.09 / 24.07 ms │            23.78 / 24.03 ±0.26 / 24.50 ms │ no change │
│ QQuery 15 │ 30.82 / 31.01 ±0.17 / 31.24 ms │            31.23 / 31.70 ±0.64 / 32.95 ms │ no change │
│ QQuery 16 │ 13.32 / 13.43 ±0.06 / 13.50 ms │            13.65 / 13.76 ±0.09 / 13.88 ms │ no change │
│ QQuery 17 │ 67.48 / 68.76 ±0.86 / 69.77 ms │            67.94 / 69.79 ±1.25 / 71.12 ms │ no change │
│ QQuery 18 │ 64.30 / 67.62 ±2.46 / 71.32 ms │            63.70 / 66.95 ±2.25 / 69.28 ms │ no change │
│ QQuery 19 │ 29.31 / 29.58 ±0.24 / 29.94 ms │            28.93 / 29.27 ±0.31 / 29.75 ms │ no change │
│ QQuery 20 │ 33.77 / 34.22 ±0.45 / 34.92 ms │            33.69 / 34.14 ±0.29 / 34.58 ms │ no change │
│ QQuery 21 │ 52.99 / 53.95 ±0.95 / 55.74 ms │            52.27 / 52.66 ±0.26 / 52.99 ms │ no change │
│ QQuery 22 │ 15.63 / 16.07 ±0.48 / 17.01 ms │            15.34 / 15.64 ±0.27 / 15.97 ms │ no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 769.72ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 768.10ms │
│ Average Time (HEAD)                                      │  34.99ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  34.91ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 468.9 MiB
CPU user 19.9s
CPU sys 1.6s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 474.7 MiB
CPU user 19.9s
CPU sys 1.7s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.29 ms │                                   5.21 ms │    no change │
│ QQuery 2  │   75.84 ms │                                  75.41 ms │    no change │
│ QQuery 3  │   28.50 ms │                                  28.40 ms │    no change │
│ QQuery 4  │  556.66 ms │                                 553.78 ms │    no change │
│ QQuery 5  │   49.80 ms │                                  49.55 ms │    no change │
│ QQuery 6  │   32.51 ms │                                  32.40 ms │    no change │
│ QQuery 7  │   69.95 ms │                                  69.54 ms │    no change │
│ QQuery 8  │   37.75 ms │                                  37.29 ms │    no change │
│ QQuery 9  │   50.73 ms │                                  52.14 ms │    no change │
│ QQuery 10 │   57.90 ms │                                  58.02 ms │    no change │
│ QQuery 11 │  362.33 ms │                                 359.79 ms │    no change │
│ QQuery 12 │   25.17 ms │                                  25.19 ms │    no change │
│ QQuery 13 │  110.62 ms │                                 110.52 ms │    no change │
│ QQuery 14 │  374.10 ms │                                 373.21 ms │    no change │
│ QQuery 15 │   50.83 ms │                                  50.87 ms │    no change │
│ QQuery 16 │    5.86 ms │                                   5.90 ms │    no change │
│ QQuery 17 │   76.10 ms │                                  75.50 ms │    no change │
│ QQuery 18 │  102.01 ms │                                 101.34 ms │    no change │
│ QQuery 19 │   40.81 ms │                                  40.75 ms │    no change │
│ QQuery 20 │   30.33 ms │                                  30.55 ms │    no change │
│ QQuery 21 │   16.28 ms │                                  16.22 ms │    no change │
│ QQuery 22 │   67.67 ms │                                  68.02 ms │    no change │
│ QQuery 23 │  285.96 ms │                                 287.04 ms │    no change │
│ QQuery 24 │  184.77 ms │                                 184.05 ms │    no change │
│ QQuery 25 │  102.64 ms │                                 102.38 ms │    no change │
│ QQuery 26 │   43.71 ms │                                  43.68 ms │    no change │
│ QQuery 27 │    5.20 ms │                                   5.00 ms │    no change │
│ QQuery 28 │   60.54 ms │                                  61.36 ms │    no change │
│ QQuery 29 │   91.78 ms │                                  91.49 ms │    no change │
│ QQuery 30 │   35.12 ms │                                  35.24 ms │    no change │
│ QQuery 31 │  107.40 ms │                                 107.88 ms │    no change │
│ QQuery 32 │   19.26 ms │                                  19.11 ms │    no change │
│ QQuery 33 │   36.29 ms │                                  35.92 ms │    no change │
│ QQuery 34 │    5.52 ms │                                   5.61 ms │    no change │
│ QQuery 35 │   69.31 ms │                                  69.02 ms │    no change │
│ QQuery 36 │    5.64 ms │                                   5.82 ms │    no change │
│ QQuery 37 │    7.29 ms │                                   7.29 ms │    no change │
│ QQuery 38 │   62.96 ms │                                  62.79 ms │    no change │
│ QQuery 39 │   71.04 ms │                                  71.25 ms │    no change │
│ QQuery 40 │   21.82 ms │                                  21.88 ms │    no change │
│ QQuery 41 │   13.99 ms │                                  14.15 ms │    no change │
│ QQuery 42 │   22.95 ms │                                  23.18 ms │    no change │
│ QQuery 43 │    4.64 ms │                                   4.62 ms │    no change │
│ QQuery 44 │    5.40 ms │                                   5.35 ms │    no change │
│ QQuery 45 │   33.99 ms │                                  34.36 ms │    no change │
│ QQuery 46 │   10.95 ms │                                  11.18 ms │    no change │
│ QQuery 47 │  196.60 ms │                                 198.20 ms │    no change │
│ QQuery 48 │   89.11 ms │                                  89.83 ms │    no change │
│ QQuery 49 │   67.96 ms │                                  68.29 ms │    no change │
│ QQuery 50 │   55.88 ms │                                  56.48 ms │    no change │
│ QQuery 51 │   91.68 ms │                                  92.97 ms │    no change │
│ QQuery 52 │   23.55 ms │                                  23.35 ms │    no change │
│ QQuery 53 │   28.50 ms │                                  28.64 ms │    no change │
│ QQuery 54 │   52.15 ms │                                  52.18 ms │    no change │
│ QQuery 55 │   22.88 ms │                                  23.00 ms │    no change │
│ QQuery 56 │   37.49 ms │                                  37.58 ms │    no change │
│ QQuery 57 │  149.71 ms │                                 149.09 ms │    no change │
│ QQuery 58 │  102.22 ms │                                 102.30 ms │    no change │
│ QQuery 59 │  105.91 ms │                                 106.17 ms │    no change │
│ QQuery 60 │   38.09 ms │                                  37.69 ms │    no change │
│ QQuery 61 │    8.85 ms │                                   8.92 ms │    no change │
│ QQuery 62 │   37.32 ms │                                  36.93 ms │    no change │
│ QQuery 63 │   29.14 ms │                                  28.87 ms │    no change │
│ QQuery 64 │  348.15 ms │                                 348.95 ms │    no change │
│ QQuery 65 │   74.80 ms │                                  74.77 ms │    no change │
│ QQuery 66 │   70.95 ms │                                  70.90 ms │    no change │
│ QQuery 67 │  216.17 ms │                                 217.07 ms │    no change │
│ QQuery 68 │    6.38 ms │                                   6.36 ms │    no change │
│ QQuery 69 │   53.06 ms │                                  52.74 ms │    no change │
│ QQuery 70 │   95.14 ms │                                  94.92 ms │    no change │
│ QQuery 71 │   35.06 ms │                                  34.70 ms │    no change │
│ QQuery 72 │ 1517.40 ms │                                1612.13 ms │ 1.06x slower │
│ QQuery 73 │    5.09 ms │                                   5.24 ms │    no change │
│ QQuery 74 │  240.96 ms │                                 243.46 ms │    no change │
│ QQuery 75 │  137.43 ms │                                 137.46 ms │    no change │
│ QQuery 76 │   33.51 ms │                                  33.91 ms │    no change │
│ QQuery 77 │   58.35 ms │                                  58.71 ms │    no change │
│ QQuery 78 │  251.70 ms │                                 257.23 ms │    no change │
│ QQuery 79 │   66.43 ms │                                  66.16 ms │    no change │
│ QQuery 80 │   91.39 ms │                                  92.75 ms │    no change │
│ QQuery 81 │   27.43 ms │                                  27.53 ms │    no change │
│ QQuery 82 │   16.55 ms │                                  16.72 ms │    no change │
│ QQuery 83 │   31.76 ms │                                  31.55 ms │    no change │
│ QQuery 84 │   30.01 ms │                                  29.95 ms │    no change │
│ QQuery 85 │   93.70 ms │                                  94.05 ms │    no change │
│ QQuery 86 │   22.96 ms │                                  23.44 ms │    no change │
│ QQuery 87 │   63.05 ms │                                  63.99 ms │    no change │
│ QQuery 88 │   58.21 ms │                                  58.67 ms │    no change │
│ QQuery 89 │   35.13 ms │                                  34.94 ms │    no change │
│ QQuery 90 │   15.72 ms │                                  15.92 ms │    no change │
│ QQuery 91 │   38.55 ms │                                  39.16 ms │    no change │
│ QQuery 92 │   28.71 ms │                                  27.93 ms │    no change │
│ QQuery 93 │   48.21 ms │                                  48.42 ms │    no change │
│ QQuery 94 │   37.42 ms │                                  36.86 ms │    no change │
│ QQuery 95 │   77.98 ms │                                  78.18 ms │    no change │
│ QQuery 96 │   23.17 ms │                                  23.35 ms │    no change │
│ QQuery 97 │   50.30 ms │                                  49.80 ms │    no change │
│ QQuery 98 │   37.30 ms │                                  37.09 ms │    no change │
│ QQuery 99 │   55.21 ms │                                  55.05 ms │    no change │
└───────────┴────────────┴───────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 8765.61ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 8869.79ms │
│ Average Time (HEAD)                                      │   88.54ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   89.59ms │
│ Queries Faster                                           │         0 │
│ Queries Slower                                           │         1 │
│ Queries with No Change                                   │        98 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.29 / 5.94 ±1.21 / 8.37 ms │               5.21 / 5.82 ±1.07 / 7.97 ms │     no change │
│ QQuery 2  │        75.84 / 75.97 ±0.15 / 76.25 ms │            75.41 / 75.84 ±0.30 / 76.34 ms │     no change │
│ QQuery 3  │        28.50 / 28.63 ±0.12 / 28.85 ms │            28.40 / 28.72 ±0.31 / 29.12 ms │     no change │
│ QQuery 4  │     556.66 / 560.61 ±3.29 / 566.44 ms │         553.78 / 559.33 ±5.06 / 568.87 ms │     no change │
│ QQuery 5  │        49.80 / 50.13 ±0.21 / 50.39 ms │            49.55 / 49.98 ±0.35 / 50.54 ms │     no change │
│ QQuery 6  │        32.51 / 33.05 ±0.44 / 33.61 ms │            32.40 / 32.88 ±0.50 / 33.75 ms │     no change │
│ QQuery 7  │        69.95 / 71.89 ±2.30 / 76.35 ms │            69.54 / 70.99 ±1.83 / 74.52 ms │     no change │
│ QQuery 8  │        37.75 / 38.13 ±0.43 / 38.96 ms │            37.29 / 37.72 ±0.37 / 38.29 ms │     no change │
│ QQuery 9  │        50.73 / 52.96 ±2.07 / 55.87 ms │            52.14 / 53.66 ±0.97 / 55.07 ms │     no change │
│ QQuery 10 │        57.90 / 58.33 ±0.53 / 59.34 ms │            58.02 / 58.39 ±0.20 / 58.59 ms │     no change │
│ QQuery 11 │     362.33 / 367.39 ±3.93 / 372.72 ms │         359.79 / 365.81 ±3.63 / 370.17 ms │     no change │
│ QQuery 12 │        25.17 / 25.43 ±0.22 / 25.75 ms │            25.19 / 25.56 ±0.36 / 26.21 ms │     no change │
│ QQuery 13 │     110.62 / 112.36 ±2.04 / 116.24 ms │         110.52 / 111.92 ±0.81 / 112.74 ms │     no change │
│ QQuery 14 │     374.10 / 376.23 ±2.30 / 380.60 ms │         373.21 / 375.99 ±1.69 / 378.40 ms │     no change │
│ QQuery 15 │        50.83 / 51.13 ±0.28 / 51.65 ms │            50.87 / 51.39 ±0.30 / 51.74 ms │     no change │
│ QQuery 16 │           5.86 / 5.96 ±0.13 / 6.22 ms │               5.90 / 6.07 ±0.28 / 6.63 ms │     no change │
│ QQuery 17 │        76.10 / 76.35 ±0.25 / 76.81 ms │            75.50 / 76.04 ±0.36 / 76.49 ms │     no change │
│ QQuery 18 │     102.01 / 104.54 ±2.03 / 108.09 ms │         101.34 / 103.33 ±2.09 / 107.37 ms │     no change │
│ QQuery 19 │        40.81 / 41.27 ±0.48 / 42.13 ms │            40.75 / 41.36 ±0.42 / 42.05 ms │     no change │
│ QQuery 20 │        30.33 / 30.80 ±0.24 / 31.01 ms │            30.55 / 30.81 ±0.18 / 31.05 ms │     no change │
│ QQuery 21 │        16.28 / 16.48 ±0.29 / 17.04 ms │            16.22 / 16.45 ±0.23 / 16.89 ms │     no change │
│ QQuery 22 │        67.67 / 69.70 ±2.19 / 73.72 ms │            68.02 / 69.45 ±1.34 / 71.99 ms │     no change │
│ QQuery 23 │     285.96 / 289.25 ±2.22 / 292.43 ms │         287.04 / 289.93 ±2.05 / 293.25 ms │     no change │
│ QQuery 24 │     184.77 / 187.18 ±1.97 / 190.64 ms │         184.05 / 187.65 ±2.54 / 191.43 ms │     no change │
│ QQuery 25 │     102.64 / 105.64 ±4.47 / 114.36 ms │         102.38 / 107.12 ±6.79 / 120.48 ms │     no change │
│ QQuery 26 │        43.71 / 44.10 ±0.31 / 44.66 ms │            43.68 / 44.28 ±0.50 / 44.91 ms │     no change │
│ QQuery 27 │           5.20 / 5.29 ±0.12 / 5.53 ms │               5.00 / 5.25 ±0.15 / 5.47 ms │     no change │
│ QQuery 28 │        60.54 / 61.44 ±0.63 / 62.19 ms │            61.36 / 61.90 ±0.37 / 62.46 ms │     no change │
│ QQuery 29 │        91.78 / 94.12 ±2.67 / 98.43 ms │            91.49 / 93.53 ±2.49 / 97.80 ms │     no change │
│ QQuery 30 │        35.12 / 35.51 ±0.31 / 36.03 ms │            35.24 / 35.52 ±0.23 / 35.85 ms │     no change │
│ QQuery 31 │     107.40 / 107.88 ±0.65 / 109.16 ms │         107.88 / 108.57 ±0.52 / 109.40 ms │     no change │
│ QQuery 32 │        19.26 / 19.58 ±0.36 / 20.26 ms │            19.11 / 19.39 ±0.32 / 20.02 ms │     no change │
│ QQuery 33 │        36.29 / 38.68 ±3.52 / 45.63 ms │            35.92 / 37.09 ±1.11 / 39.21 ms │     no change │
│ QQuery 34 │           5.52 / 5.67 ±0.23 / 6.12 ms │               5.61 / 5.79 ±0.26 / 6.30 ms │     no change │
│ QQuery 35 │        69.31 / 69.67 ±0.39 / 70.43 ms │            69.02 / 69.16 ±0.10 / 69.31 ms │     no change │
│ QQuery 36 │           5.64 / 5.79 ±0.20 / 6.19 ms │               5.82 / 5.98 ±0.22 / 6.41 ms │     no change │
│ QQuery 37 │           7.29 / 7.57 ±0.19 / 7.88 ms │               7.29 / 7.39 ±0.12 / 7.63 ms │     no change │
│ QQuery 38 │        62.96 / 63.48 ±0.41 / 64.08 ms │            62.79 / 63.76 ±0.70 / 64.85 ms │     no change │
│ QQuery 39 │        71.04 / 72.78 ±2.76 / 78.28 ms │            71.25 / 73.46 ±2.52 / 78.29 ms │     no change │
│ QQuery 40 │        21.82 / 22.23 ±0.33 / 22.62 ms │            21.88 / 22.54 ±0.61 / 23.34 ms │     no change │
│ QQuery 41 │        13.99 / 14.10 ±0.14 / 14.37 ms │            14.15 / 14.31 ±0.20 / 14.69 ms │     no change │
│ QQuery 42 │        22.95 / 23.50 ±0.52 / 24.45 ms │            23.18 / 23.90 ±0.57 / 24.55 ms │     no change │
│ QQuery 43 │           4.64 / 4.75 ±0.13 / 4.99 ms │               4.62 / 4.74 ±0.14 / 5.00 ms │     no change │
│ QQuery 44 │           5.40 / 5.43 ±0.04 / 5.52 ms │               5.35 / 5.41 ±0.07 / 5.54 ms │     no change │
│ QQuery 45 │        33.99 / 34.20 ±0.25 / 34.62 ms │            34.36 / 34.88 ±0.45 / 35.43 ms │     no change │
│ QQuery 46 │        10.95 / 11.05 ±0.09 / 11.20 ms │            11.18 / 11.29 ±0.09 / 11.41 ms │     no change │
│ QQuery 47 │     196.60 / 199.86 ±3.08 / 204.94 ms │         198.20 / 200.68 ±2.92 / 205.93 ms │     no change │
│ QQuery 48 │        89.11 / 91.75 ±3.07 / 97.62 ms │            89.83 / 91.37 ±1.82 / 94.83 ms │     no change │
│ QQuery 49 │        67.96 / 68.89 ±0.84 / 70.40 ms │            68.29 / 69.51 ±0.96 / 70.97 ms │     no change │
│ QQuery 50 │        55.88 / 56.06 ±0.19 / 56.40 ms │            56.48 / 56.99 ±0.52 / 57.90 ms │     no change │
│ QQuery 51 │       91.68 / 94.76 ±3.31 / 101.16 ms │            92.97 / 95.33 ±2.12 / 98.68 ms │     no change │
│ QQuery 52 │        23.55 / 23.81 ±0.20 / 24.09 ms │            23.35 / 23.71 ±0.22 / 23.98 ms │     no change │
│ QQuery 53 │        28.50 / 28.76 ±0.16 / 28.99 ms │            28.64 / 29.09 ±0.24 / 29.34 ms │     no change │
│ QQuery 54 │        52.15 / 52.42 ±0.24 / 52.75 ms │            52.18 / 52.36 ±0.14 / 52.55 ms │     no change │
│ QQuery 55 │        22.88 / 24.94 ±3.38 / 31.69 ms │            23.00 / 25.39 ±3.49 / 32.26 ms │     no change │
│ QQuery 56 │        37.49 / 38.36 ±0.69 / 39.20 ms │            37.58 / 38.09 ±0.36 / 38.62 ms │     no change │
│ QQuery 57 │     149.71 / 150.75 ±0.73 / 151.97 ms │         149.09 / 151.56 ±1.81 / 154.65 ms │     no change │
│ QQuery 58 │     102.22 / 103.78 ±1.75 / 107.16 ms │         102.30 / 104.52 ±2.43 / 109.20 ms │     no change │
│ QQuery 59 │     105.91 / 107.01 ±0.82 / 107.84 ms │         106.17 / 107.53 ±1.49 / 110.32 ms │     no change │
│ QQuery 60 │        38.09 / 38.55 ±0.43 / 39.24 ms │            37.69 / 38.59 ±0.66 / 39.60 ms │     no change │
│ QQuery 61 │           8.85 / 9.06 ±0.30 / 9.65 ms │               8.92 / 9.08 ±0.20 / 9.48 ms │     no change │
│ QQuery 62 │        37.32 / 37.74 ±0.37 / 38.39 ms │            36.93 / 37.70 ±0.49 / 38.25 ms │     no change │
│ QQuery 63 │        29.14 / 29.26 ±0.10 / 29.39 ms │            28.87 / 29.25 ±0.47 / 30.06 ms │     no change │
│ QQuery 64 │     348.15 / 356.74 ±8.66 / 373.00 ms │         348.95 / 355.56 ±4.02 / 360.25 ms │     no change │
│ QQuery 65 │        74.80 / 75.75 ±0.62 / 76.50 ms │            74.77 / 75.25 ±0.75 / 76.75 ms │     no change │
│ QQuery 66 │        70.95 / 73.47 ±4.12 / 81.67 ms │            70.90 / 73.09 ±2.74 / 78.40 ms │     no change │
│ QQuery 67 │     216.17 / 219.69 ±3.34 / 225.89 ms │         217.07 / 225.05 ±5.47 / 231.46 ms │     no change │
│ QQuery 68 │           6.38 / 6.54 ±0.25 / 7.03 ms │               6.36 / 6.55 ±0.29 / 7.12 ms │     no change │
│ QQuery 69 │        53.06 / 53.37 ±0.18 / 53.58 ms │            52.74 / 53.04 ±0.21 / 53.34 ms │     no change │
│ QQuery 70 │        95.14 / 95.61 ±0.28 / 95.94 ms │            94.92 / 95.66 ±1.25 / 98.14 ms │     no change │
│ QQuery 71 │        35.06 / 38.47 ±4.56 / 47.37 ms │            34.70 / 35.44 ±0.72 / 36.82 ms │ +1.09x faster │
│ QQuery 72 │ 1517.40 / 1619.02 ±69.41 / 1696.80 ms │     1612.13 / 1659.09 ±36.19 / 1694.29 ms │     no change │
│ QQuery 73 │           5.09 / 5.26 ±0.25 / 5.74 ms │               5.24 / 5.40 ±0.24 / 5.86 ms │     no change │
│ QQuery 74 │     240.96 / 244.95 ±3.13 / 249.80 ms │         243.46 / 245.98 ±2.47 / 250.58 ms │     no change │
│ QQuery 75 │     137.43 / 140.11 ±2.89 / 144.97 ms │         137.46 / 138.32 ±0.74 / 139.60 ms │     no change │
│ QQuery 76 │        33.51 / 34.17 ±0.57 / 35.02 ms │            33.91 / 34.17 ±0.18 / 34.39 ms │     no change │
│ QQuery 77 │        58.35 / 58.73 ±0.24 / 59.11 ms │            58.71 / 59.33 ±0.46 / 59.93 ms │     no change │
│ QQuery 78 │    251.70 / 263.29 ±12.49 / 285.48 ms │        257.23 / 273.17 ±16.14 / 296.46 ms │     no change │
│ QQuery 79 │        66.43 / 68.85 ±2.80 / 74.04 ms │            66.16 / 66.82 ±0.54 / 67.80 ms │     no change │
│ QQuery 80 │        91.39 / 93.90 ±1.63 / 95.86 ms │            92.75 / 93.64 ±0.94 / 95.36 ms │     no change │
│ QQuery 81 │        27.43 / 27.96 ±0.48 / 28.73 ms │            27.53 / 30.40 ±4.71 / 39.81 ms │  1.09x slower │
│ QQuery 82 │        16.55 / 17.35 ±0.97 / 19.26 ms │            16.72 / 17.03 ±0.31 / 17.47 ms │     no change │
│ QQuery 83 │        31.76 / 34.16 ±2.99 / 39.19 ms │            31.55 / 31.70 ±0.18 / 32.05 ms │ +1.08x faster │
│ QQuery 84 │        30.01 / 30.51 ±0.52 / 31.40 ms │            29.95 / 30.14 ±0.16 / 30.41 ms │     no change │
│ QQuery 85 │        93.70 / 94.62 ±0.97 / 95.84 ms │           94.05 / 97.68 ±6.33 / 110.30 ms │     no change │
│ QQuery 86 │        22.96 / 23.37 ±0.39 / 24.10 ms │            23.44 / 23.75 ±0.29 / 24.29 ms │     no change │
│ QQuery 87 │        63.05 / 63.91 ±0.52 / 64.64 ms │            63.99 / 64.38 ±0.24 / 64.70 ms │     no change │
│ QQuery 88 │        58.21 / 58.91 ±0.54 / 59.70 ms │            58.67 / 59.25 ±0.39 / 59.72 ms │     no change │
│ QQuery 89 │        35.13 / 35.60 ±0.32 / 36.00 ms │            34.94 / 35.61 ±0.49 / 36.39 ms │     no change │
│ QQuery 90 │        15.72 / 15.89 ±0.15 / 16.16 ms │            15.92 / 16.17 ±0.25 / 16.63 ms │     no change │
│ QQuery 91 │        38.55 / 38.88 ±0.28 / 39.38 ms │            39.16 / 41.31 ±3.90 / 49.10 ms │  1.06x slower │
│ QQuery 92 │        28.71 / 30.55 ±2.08 / 33.37 ms │            27.93 / 28.57 ±0.49 / 29.40 ms │ +1.07x faster │
│ QQuery 93 │        48.21 / 49.28 ±0.55 / 49.73 ms │            48.42 / 49.23 ±0.87 / 50.92 ms │     no change │
│ QQuery 94 │        37.42 / 37.82 ±0.27 / 38.14 ms │            36.86 / 37.34 ±0.42 / 37.93 ms │     no change │
│ QQuery 95 │        77.98 / 78.66 ±0.40 / 79.07 ms │            78.18 / 79.67 ±1.13 / 81.61 ms │     no change │
│ QQuery 96 │        23.17 / 23.83 ±0.88 / 25.52 ms │            23.35 / 23.79 ±0.60 / 24.90 ms │     no change │
│ QQuery 97 │        50.30 / 52.15 ±1.98 / 55.86 ms │            49.80 / 50.55 ±0.52 / 51.17 ms │     no change │
│ QQuery 98 │        37.30 / 37.64 ±0.37 / 38.27 ms │            37.09 / 37.64 ±0.45 / 38.21 ms │     no change │
│ QQuery 99 │        55.21 / 55.56 ±0.22 / 55.85 ms │            55.05 / 56.99 ±2.10 / 61.07 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 8992.53ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 9047.89ms │
│ Average Time (HEAD)                                      │   90.83ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   91.39ms │
│ Queries Faster                                           │         3 │
│ Queries Slower                                           │         2 │
│ Queries with No Change                                   │        94 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 1.9 GiB
Avg memory 1.2 GiB
CPU user 182.3s
CPU sys 5.4s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 1.9 GiB
Avg memory 1.3 GiB
CPU user 185.9s
CPU sys 5.3s
Peak spill 0 B

File an issue against this benchmark runner

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark tpcds

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6758911930-4741-5frzd 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark dictionary_group_values
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

group                                                                             HEAD                                    rich-t-kid_single-dictionary-group-values
-----                                                                             ----                                    -----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.90    479.0±8.00µs 130.5 MElem/sec    1.00    252.5±0.62µs 247.6 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.95    413.5±5.83µs 151.2 MElem/sec    1.00    211.5±1.16µs 295.5 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.98    441.5±2.67µs 141.5 MElem/sec    1.00    222.9±1.20µs 280.3 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.81      5.9±0.02ms 10.5 MElem/sec     1.00      3.3±0.01ms 19.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         2.02    435.0±2.59µs 143.7 MElem/sec    1.00    215.0±1.39µs 290.7 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.60    111.7±0.62µs 70.0 MElem/sec     1.00     69.6±0.42µs 112.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.96     55.6±1.16µs 140.4 MElem/sec    1.00     28.4±0.57µs 275.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.81     72.6±0.58µs 107.6 MElem/sec    1.00     40.1±0.36µs 194.7 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.99     60.5±0.52µs 129.1 MElem/sec    1.00     30.4±0.16µs 257.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.42    520.9±5.71µs 15.0 MElem/sec     1.00    366.2±1.24µs 21.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.63      2.2±0.01ms 113.9 MElem/sec    1.00   1343.8±2.67µs 186.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.67      2.0±0.00ms 122.6 MElem/sec    1.00   1221.3±4.04µs 204.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.70      2.1±0.00ms 117.4 MElem/sec    1.00   1252.7±1.48µs 199.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.72     18.7±0.05ms 13.3 MElem/sec     1.00     10.9±0.10ms 22.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.72      2.1±0.00ms 118.4 MElem/sec    1.00   1228.9±3.56µs 203.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.46   383.8±19.71µs 81.4 MElem/sec     1.00    262.4±1.57µs 119.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.77    248.5±1.55µs 125.8 MElem/sec    1.00    140.7±0.47µs 222.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.66    286.9±2.68µs 108.9 MElem/sec    1.00    172.5±0.71µs 181.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.92  282.1±204.28µs 110.8 MElem/sec    1.00    146.6±0.35µs 213.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.27   1289.2±4.65µs 24.2 MElem/sec     1.00  1014.9±11.39µs 30.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       3.82   1337.8±4.00µs 46.7 MElem/sec     1.00    350.6±1.55µs 178.2 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         4.04   1210.1±3.54µs 51.6 MElem/sec     1.00    299.3±0.71µs 208.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        4.10   1281.1±3.11µs 48.8 MElem/sec     1.00    312.4±1.35µs 200.1 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      2.08      7.4±0.02ms  8.5 MElem/sec     1.00      3.5±0.04ms 17.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         4.07   1230.0±4.17µs 50.8 MElem/sec     1.00    302.2±0.73µs 206.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    2.42     18.0±0.03ms 13.9 MElem/sec     1.00      7.4±0.01ms 33.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     2.81      7.7±0.01ms 32.6 MElem/sec     1.00      2.7±0.01ms 91.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    2.90     57.7±1.15ms  4.3 MElem/sec     1.00     19.9±0.14ms 12.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     2.47     13.6±0.02ms 18.3 MElem/sec     1.00      5.5±0.01ms 45.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      2.63      2.8±0.01ms 88.8 MElem/sec     1.00   1071.6±3.57µs 233.3 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.66      2.3±0.01ms 107.1 MElem/sec    1.00   1408.6±3.78µs 177.5 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.68      2.0±0.00ms 124.9 MElem/sec    1.00   1190.8±1.89µs 209.9 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.72      2.1±0.00ms 117.3 MElem/sec    1.00   1242.2±3.12µs 201.3 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.62     25.6±0.05ms  9.7 MElem/sec     1.00     15.8±0.06ms 15.8 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.73      2.1±0.00ms 120.4 MElem/sec    1.00   1201.7±3.35µs 208.0 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric Value
Wall time 595.1s
Peak memory 901.3 MiB
Avg memory 68.7 MiB
CPU user 403.8s
CPU sys 23.0s
Peak spill 0 B

dictionary_group_values — branch

Metric Value
Wall time 605.1s
Peak memory 551.7 MiB
Avg memory 332.9 MiB
CPU user 454.4s
CPU sys 0.5s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.12 ms │                                   1.13 ms │     no change │
│ QQuery 1  │   11.46 ms │                                  11.25 ms │     no change │
│ QQuery 2  │   34.64 ms │                                  34.14 ms │     no change │
│ QQuery 3  │   30.37 ms │                                  30.38 ms │     no change │
│ QQuery 4  │  238.87 ms │                                 229.63 ms │     no change │
│ QQuery 5  │  818.99 ms │                                 732.96 ms │ +1.12x faster │
│ QQuery 6  │    1.19 ms │                                   1.16 ms │     no change │
│ QQuery 7  │   12.72 ms │                                  12.57 ms │     no change │
│ QQuery 8  │  329.25 ms │                                 327.13 ms │     no change │
│ QQuery 9  │  471.07 ms │                                 466.33 ms │     no change │
│ QQuery 10 │  202.13 ms │                                 199.76 ms │     no change │
│ QQuery 11 │  212.92 ms │                                 210.23 ms │     no change │
│ QQuery 12 │  800.06 ms │                                 735.89 ms │ +1.09x faster │
│ QQuery 13 │ 1064.50 ms │                                1068.29 ms │     no change │
│ QQuery 14 │  847.25 ms │                                 839.84 ms │     no change │
│ QQuery 15 │  285.13 ms │                                 285.26 ms │     no change │
│ QQuery 16 │ 1275.28 ms │                                1286.29 ms │     no change │
│ QQuery 17 │ 1252.91 ms │                                1251.74 ms │     no change │
│ QQuery 18 │ 1949.89 ms │                                1992.60 ms │     no change │
│ QQuery 19 │   27.60 ms │                                  27.44 ms │     no change │
│ QQuery 20 │ 1041.23 ms │                                1047.91 ms │     no change │
│ QQuery 21 │ 1368.43 ms │                                1354.87 ms │     no change │
│ QQuery 22 │ 3198.37 ms │                                3210.91 ms │     no change │
│ QQuery 23 │ 9219.62 ms │                                9151.99 ms │     no change │
│ QQuery 24 │   95.57 ms │                                  96.31 ms │     no change │
│ QQuery 25 │  351.63 ms │                                 352.15 ms │     no change │
│ QQuery 26 │   98.52 ms │                                  97.70 ms │     no change │
│ QQuery 27 │ 1140.79 ms │                                1143.49 ms │     no change │
│ QQuery 28 │ 2529.58 ms │                                2440.81 ms │     no change │
│ QQuery 29 │   36.88 ms │                                  36.98 ms │     no change │
│ QQuery 30 │  537.59 ms │                                 544.34 ms │     no change │
│ QQuery 31 │  536.64 ms │                                 532.44 ms │     no change │
│ QQuery 32 │ 1072.10 ms │                                1062.65 ms │     no change │
│ QQuery 33 │ 3574.54 ms │                                3328.81 ms │ +1.07x faster │
│ QQuery 34 │ 3545.18 ms │                                3384.84 ms │     no change │
│ QQuery 35 │  289.14 ms │                                 290.80 ms │     no change │
│ QQuery 36 │  132.82 ms │                                 122.75 ms │ +1.08x faster │
│ QQuery 37 │   96.32 ms │                                  94.51 ms │     no change │
│ QQuery 38 │   78.79 ms │                                  80.78 ms │     no change │
│ QQuery 39 │  293.27 ms │                                 281.55 ms │     no change │
│ QQuery 40 │   13.18 ms │                                  12.90 ms │     no change │
│ QQuery 41 │   12.50 ms │                                  12.41 ms │     no change │
│ QQuery 42 │   11.73 ms │                                  12.23 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 39141.75ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 38438.14ms │
│ Average Time (HEAD)                                      │   910.27ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   893.91ms │
│ Queries Faster                                           │          4 │
│ Queries Slower                                           │          0 │
│ Queries with No Change                                   │         39 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                   HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │            1.12 / 1.76 ±1.14 / 4.03 ms │               1.13 / 1.69 ±1.05 / 3.78 ms │     no change │
│ QQuery 1  │         11.46 / 11.55 ±0.09 / 11.69 ms │            11.25 / 11.47 ±0.14 / 11.68 ms │     no change │
│ QQuery 2  │         34.64 / 35.06 ±0.46 / 35.89 ms │            34.14 / 34.48 ±0.23 / 34.73 ms │     no change │
│ QQuery 3  │         30.37 / 31.03 ±0.79 / 32.51 ms │            30.38 / 30.58 ±0.16 / 30.83 ms │     no change │
│ QQuery 4  │      238.87 / 243.18 ±2.47 / 245.83 ms │         229.63 / 236.43 ±3.92 / 241.02 ms │     no change │
│ QQuery 5  │      818.99 / 824.17 ±4.45 / 830.76 ms │         732.96 / 741.96 ±4.90 / 747.31 ms │ +1.11x faster │
│ QQuery 6  │            1.19 / 1.36 ±0.24 / 1.83 ms │               1.16 / 1.32 ±0.24 / 1.78 ms │     no change │
│ QQuery 7  │         12.72 / 12.95 ±0.26 / 13.39 ms │            12.57 / 12.70 ±0.08 / 12.81 ms │     no change │
│ QQuery 8  │      329.25 / 334.33 ±3.16 / 338.42 ms │         327.13 / 332.62 ±4.09 / 337.76 ms │     no change │
│ QQuery 9  │      471.07 / 474.75 ±2.97 / 478.89 ms │        466.33 / 478.24 ±17.59 / 513.24 ms │     no change │
│ QQuery 10 │      202.13 / 202.80 ±0.51 / 203.38 ms │         199.76 / 201.36 ±2.48 / 206.29 ms │     no change │
│ QQuery 11 │      212.92 / 214.43 ±0.98 / 215.40 ms │         210.23 / 210.78 ±0.37 / 211.34 ms │     no change │
│ QQuery 12 │     800.06 / 818.15 ±12.91 / 833.81 ms │        735.89 / 759.11 ±14.16 / 778.41 ms │ +1.08x faster │
│ QQuery 13 │  1064.50 / 1107.77 ±25.19 / 1137.07 ms │     1068.29 / 1091.97 ±25.87 / 1141.53 ms │     no change │
│ QQuery 14 │     847.25 / 861.15 ±15.45 / 889.87 ms │        839.84 / 854.71 ±12.01 / 873.35 ms │     no change │
│ QQuery 15 │      285.13 / 291.14 ±8.48 / 307.77 ms │         285.26 / 291.57 ±4.31 / 298.69 ms │     no change │
│ QQuery 16 │  1275.28 / 1298.63 ±27.13 / 1346.60 ms │     1286.29 / 1316.54 ±21.34 / 1346.03 ms │     no change │
│ QQuery 17 │   1252.91 / 1259.91 ±5.71 / 1268.34 ms │     1251.74 / 1274.11 ±25.31 / 1321.09 ms │     no change │
│ QQuery 18 │  1949.89 / 1997.36 ±28.56 / 2031.52 ms │     1992.60 / 2032.08 ±22.77 / 2056.47 ms │     no change │
│ QQuery 19 │         27.60 / 29.79 ±3.35 / 36.42 ms │           27.44 / 38.89 ±19.23 / 77.13 ms │  1.31x slower │
│ QQuery 20 │   1041.23 / 1049.14 ±6.87 / 1058.97 ms │      1047.91 / 1060.39 ±8.93 / 1070.45 ms │     no change │
│ QQuery 21 │   1368.43 / 1370.11 ±2.22 / 1374.39 ms │      1354.87 / 1371.67 ±9.65 / 1383.73 ms │     no change │
│ QQuery 22 │  3198.37 / 3239.37 ±25.40 / 3278.37 ms │     3210.91 / 3267.33 ±47.02 / 3331.63 ms │     no change │
│ QQuery 23 │  9219.62 / 9356.00 ±81.98 / 9474.13 ms │    9151.99 / 9445.52 ±192.49 / 9725.91 ms │     no change │
│ QQuery 24 │      95.57 / 125.46 ±35.79 / 170.55 ms │            96.31 / 97.71 ±1.29 / 99.64 ms │ +1.28x faster │
│ QQuery 25 │     351.63 / 372.32 ±34.96 / 442.16 ms │         352.15 / 356.43 ±3.41 / 360.66 ms │     no change │
│ QQuery 26 │        98.52 / 99.45 ±0.95 / 101.20 ms │          97.70 / 100.02 ±2.54 / 104.20 ms │     no change │
│ QQuery 27 │   1140.79 / 1145.28 ±3.57 / 1150.59 ms │      1143.49 / 1150.76 ±5.93 / 1157.69 ms │     no change │
│ QQuery 28 │  2529.58 / 2552.95 ±24.14 / 2597.27 ms │     2440.81 / 2499.10 ±36.37 / 2542.89 ms │     no change │
│ QQuery 29 │         36.88 / 37.29 ±0.41 / 38.02 ms │           36.98 / 49.13 ±23.76 / 96.65 ms │  1.32x slower │
│ QQuery 30 │      537.59 / 549.16 ±7.30 / 558.78 ms │         544.34 / 550.21 ±4.10 / 557.21 ms │     no change │
│ QQuery 31 │     536.64 / 556.78 ±23.39 / 602.56 ms │        532.44 / 553.25 ±17.23 / 579.39 ms │     no change │
│ QQuery 32 │  1072.10 / 1107.18 ±28.99 / 1152.70 ms │     1062.65 / 1128.32 ±74.70 / 1272.03 ms │     no change │
│ QQuery 33 │  3574.54 / 3654.43 ±78.48 / 3804.57 ms │     3328.81 / 3438.32 ±93.84 / 3569.92 ms │ +1.06x faster │
│ QQuery 34 │ 3545.18 / 3772.51 ±121.03 / 3883.56 ms │     3384.84 / 3473.54 ±71.50 / 3541.19 ms │ +1.09x faster │
│ QQuery 35 │     289.14 / 337.25 ±66.83 / 466.10 ms │         290.80 / 302.70 ±8.41 / 315.19 ms │ +1.11x faster │
│ QQuery 36 │     132.82 / 147.63 ±13.39 / 169.55 ms │         122.75 / 134.66 ±8.26 / 145.99 ms │ +1.10x faster │
│ QQuery 37 │       96.32 / 101.39 ±5.51 / 110.49 ms │          94.51 / 100.97 ±5.58 / 110.15 ms │     no change │
│ QQuery 38 │         78.79 / 84.10 ±2.99 / 87.91 ms │            80.78 / 83.43 ±2.79 / 88.45 ms │     no change │
│ QQuery 39 │      293.27 / 304.99 ±8.32 / 318.31 ms │        281.55 / 300.76 ±14.80 / 313.85 ms │     no change │
│ QQuery 40 │         13.18 / 14.07 ±0.84 / 15.41 ms │            12.90 / 13.83 ±0.58 / 14.69 ms │     no change │
│ QQuery 41 │         12.50 / 14.38 ±2.73 / 19.69 ms │            12.41 / 13.66 ±2.17 / 18.01 ms │ +1.05x faster │
│ QQuery 42 │         11.73 / 13.73 ±3.88 / 21.49 ms │            12.23 / 14.98 ±5.29 / 25.56 ms │  1.09x slower │
└───────────┴────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 40056.27ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 39459.30ms │
│ Average Time (HEAD)                                      │   931.54ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   917.66ms │
│ Queries Faster                                           │          8 │
│ Queries Slower                                           │          3 │
│ Queries with No Change                                   │         32 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 205.1s
Peak memory 13.4 GiB
Avg memory 5.7 GiB
CPU user 1954.8s
CPU sys 131.2s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 200.0s
Peak memory 13.2 GiB
Avg memory 5.7 GiB
CPU user 1893.4s
CPU sys 133.8s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (f6ce571) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.24 ms │                                   5.28 ms │     no change │
│ QQuery 2  │   75.06 ms │                                  75.73 ms │     no change │
│ QQuery 3  │   28.44 ms │                                  28.35 ms │     no change │
│ QQuery 4  │  549.94 ms │                                 556.19 ms │     no change │
│ QQuery 5  │   49.68 ms │                                  50.11 ms │     no change │
│ QQuery 6  │   32.44 ms │                                  32.74 ms │     no change │
│ QQuery 7  │   69.46 ms │                                  70.00 ms │     no change │
│ QQuery 8  │   37.47 ms │                                  38.12 ms │     no change │
│ QQuery 9  │   51.05 ms │                                  51.52 ms │     no change │
│ QQuery 10 │   57.94 ms │                                  57.85 ms │     no change │
│ QQuery 11 │  360.80 ms │                                 363.68 ms │     no change │
│ QQuery 12 │   24.73 ms │                                  24.99 ms │     no change │
│ QQuery 13 │  110.74 ms │                                 110.75 ms │     no change │
│ QQuery 14 │  371.43 ms │                                 373.49 ms │     no change │
│ QQuery 15 │   50.06 ms │                                  50.89 ms │     no change │
│ QQuery 16 │    5.98 ms │                                   5.85 ms │     no change │
│ QQuery 17 │   75.65 ms │                                  75.63 ms │     no change │
│ QQuery 18 │  101.69 ms │                                 102.11 ms │     no change │
│ QQuery 19 │   40.51 ms │                                  40.70 ms │     no change │
│ QQuery 20 │   30.03 ms │                                  30.69 ms │     no change │
│ QQuery 21 │   16.19 ms │                                  16.27 ms │     no change │
│ QQuery 22 │   67.70 ms │                                  68.16 ms │     no change │
│ QQuery 23 │  283.88 ms │                                 288.15 ms │     no change │
│ QQuery 24 │  183.58 ms │                                 187.38 ms │     no change │
│ QQuery 25 │  102.03 ms │                                 103.04 ms │     no change │
│ QQuery 26 │   43.11 ms │                                  43.44 ms │     no change │
│ QQuery 27 │    5.17 ms │                                   5.09 ms │     no change │
│ QQuery 28 │   57.27 ms │                                  61.28 ms │  1.07x slower │
│ QQuery 29 │   91.31 ms │                                  91.07 ms │     no change │
│ QQuery 30 │   35.16 ms │                                  35.07 ms │     no change │
│ QQuery 31 │  106.96 ms │                                 107.32 ms │     no change │
│ QQuery 32 │   18.98 ms │                                  19.13 ms │     no change │
│ QQuery 33 │   35.63 ms │                                  36.14 ms │     no change │
│ QQuery 34 │    5.38 ms │                                   5.51 ms │     no change │
│ QQuery 35 │   68.18 ms │                                  68.14 ms │     no change │
│ QQuery 36 │    5.67 ms │                                   5.73 ms │     no change │
│ QQuery 37 │    7.22 ms │                                   7.22 ms │     no change │
│ QQuery 38 │   62.74 ms │                                  62.90 ms │     no change │
│ QQuery 39 │   70.66 ms │                                  70.30 ms │     no change │
│ QQuery 40 │   21.96 ms │                                  22.32 ms │     no change │
│ QQuery 41 │   14.10 ms │                                  14.00 ms │     no change │
│ QQuery 42 │   23.10 ms │                                  23.20 ms │     no change │
│ QQuery 43 │    4.53 ms │                                   4.51 ms │     no change │
│ QQuery 44 │    5.46 ms │                                   5.22 ms │     no change │
│ QQuery 45 │   34.32 ms │                                  34.57 ms │     no change │
│ QQuery 46 │   10.89 ms │                                  10.80 ms │     no change │
│ QQuery 47 │  197.79 ms │                                 197.76 ms │     no change │
│ QQuery 48 │   89.21 ms │                                  88.98 ms │     no change │
│ QQuery 49 │   68.09 ms │                                  68.50 ms │     no change │
│ QQuery 50 │   55.80 ms │                                  56.09 ms │     no change │
│ QQuery 51 │   92.25 ms │                                  93.05 ms │     no change │
│ QQuery 52 │   23.15 ms │                                  23.68 ms │     no change │
│ QQuery 53 │   28.60 ms │                                  28.57 ms │     no change │
│ QQuery 54 │   51.66 ms │                                  52.23 ms │     no change │
│ QQuery 55 │   22.83 ms │                                  22.98 ms │     no change │
│ QQuery 56 │   36.99 ms │                                  37.37 ms │     no change │
│ QQuery 57 │  149.45 ms │                                 149.14 ms │     no change │
│ QQuery 58 │  100.87 ms │                                 103.04 ms │     no change │
│ QQuery 59 │  105.17 ms │                                 106.02 ms │     no change │
│ QQuery 60 │   37.65 ms │                                  37.79 ms │     no change │
│ QQuery 61 │    8.86 ms │                                   8.88 ms │     no change │
│ QQuery 62 │   36.87 ms │                                  36.94 ms │     no change │
│ QQuery 63 │   28.64 ms │                                  28.59 ms │     no change │
│ QQuery 64 │  345.17 ms │                                 348.47 ms │     no change │
│ QQuery 65 │   74.15 ms │                                  74.63 ms │     no change │
│ QQuery 66 │   70.18 ms │                                  70.89 ms │     no change │
│ QQuery 67 │  213.81 ms │                                 216.06 ms │     no change │
│ QQuery 68 │    6.52 ms │                                   6.38 ms │     no change │
│ QQuery 69 │   52.85 ms │                                  52.87 ms │     no change │
│ QQuery 70 │   95.19 ms │                                  94.92 ms │     no change │
│ QQuery 71 │   33.99 ms │                                  34.06 ms │     no change │
│ QQuery 72 │ 1609.58 ms │                                1479.13 ms │ +1.09x faster │
│ QQuery 73 │    5.30 ms │                                   5.14 ms │     no change │
│ QQuery 74 │  242.69 ms │                                 239.08 ms │     no change │
│ QQuery 75 │  138.15 ms │                                 135.95 ms │     no change │
│ QQuery 76 │   33.85 ms │                                  33.33 ms │     no change │
│ QQuery 77 │   58.80 ms │                                  58.20 ms │     no change │
│ QQuery 78 │  255.36 ms │                                 252.07 ms │     no change │
│ QQuery 79 │   67.04 ms │                                  66.51 ms │     no change │
│ QQuery 80 │   92.54 ms │                                  92.12 ms │     no change │
│ QQuery 81 │   28.38 ms │                                  27.26 ms │     no change │
│ QQuery 82 │   17.06 ms │                                  16.61 ms │     no change │
│ QQuery 83 │   31.70 ms │                                  31.09 ms │     no change │
│ QQuery 84 │   29.96 ms │                                  29.70 ms │     no change │
│ QQuery 85 │   94.02 ms │                                  92.64 ms │     no change │
│ QQuery 86 │   23.24 ms │                                  23.37 ms │     no change │
│ QQuery 87 │   64.02 ms │                                  62.83 ms │     no change │
│ QQuery 88 │   59.44 ms │                                  59.12 ms │     no change │
│ QQuery 89 │   35.30 ms │                                  35.25 ms │     no change │
│ QQuery 90 │   15.80 ms │                                  15.67 ms │     no change │
│ QQuery 91 │   39.86 ms │                                  38.70 ms │     no change │
│ QQuery 92 │   27.94 ms │                                  27.69 ms │     no change │
│ QQuery 93 │   48.57 ms │                                  47.77 ms │     no change │
│ QQuery 94 │   36.75 ms │                                  36.89 ms │     no change │
│ QQuery 95 │   77.39 ms │                                  79.09 ms │     no change │
│ QQuery 96 │   23.22 ms │                                  23.16 ms │     no change │
│ QQuery 97 │   49.81 ms │                                  49.02 ms │     no change │
│ QQuery 98 │   37.02 ms │                                  37.20 ms │     no change │
│ QQuery 99 │   55.19 ms │                                  55.22 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 8833.25ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 8728.33ms │
│ Average Time (HEAD)                                      │   89.22ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   88.16ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │         1 │
│ Queries with No Change                                   │        97 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.24 / 5.92 ±1.19 / 8.29 ms │               5.28 / 5.90 ±1.10 / 8.10 ms │     no change │
│ QQuery 2  │        75.06 / 75.36 ±0.17 / 75.54 ms │            75.73 / 76.20 ±0.27 / 76.54 ms │     no change │
│ QQuery 3  │        28.44 / 28.52 ±0.07 / 28.62 ms │            28.35 / 28.63 ±0.14 / 28.73 ms │     no change │
│ QQuery 4  │     549.94 / 554.90 ±4.40 / 561.46 ms │         556.19 / 560.91 ±3.47 / 566.38 ms │     no change │
│ QQuery 5  │        49.68 / 50.11 ±0.60 / 51.24 ms │            50.11 / 50.62 ±0.37 / 51.16 ms │     no change │
│ QQuery 6  │        32.44 / 32.69 ±0.21 / 32.95 ms │            32.74 / 33.57 ±0.56 / 34.23 ms │     no change │
│ QQuery 7  │        69.46 / 71.53 ±2.27 / 75.78 ms │            70.00 / 71.70 ±2.32 / 76.30 ms │     no change │
│ QQuery 8  │        37.47 / 37.83 ±0.40 / 38.60 ms │            38.12 / 38.59 ±0.44 / 39.31 ms │     no change │
│ QQuery 9  │        51.05 / 52.66 ±0.96 / 53.82 ms │            51.52 / 52.99 ±1.37 / 55.47 ms │     no change │
│ QQuery 10 │        57.94 / 58.20 ±0.14 / 58.32 ms │            57.85 / 58.09 ±0.12 / 58.18 ms │     no change │
│ QQuery 11 │     360.80 / 363.02 ±2.16 / 366.75 ms │         363.68 / 367.37 ±3.30 / 372.23 ms │     no change │
│ QQuery 12 │        24.73 / 25.00 ±0.23 / 25.41 ms │            24.99 / 25.45 ±0.51 / 26.39 ms │     no change │
│ QQuery 13 │     110.74 / 113.04 ±3.51 / 119.97 ms │         110.75 / 111.46 ±0.48 / 112.01 ms │     no change │
│ QQuery 14 │     371.43 / 374.14 ±3.65 / 381.37 ms │         373.49 / 378.06 ±6.80 / 391.55 ms │     no change │
│ QQuery 15 │        50.06 / 51.02 ±0.62 / 51.89 ms │            50.89 / 51.13 ±0.21 / 51.40 ms │     no change │
│ QQuery 16 │           5.98 / 6.12 ±0.14 / 6.38 ms │               5.85 / 5.99 ±0.14 / 6.26 ms │     no change │
│ QQuery 17 │        75.65 / 76.26 ±0.59 / 77.28 ms │            75.63 / 75.92 ±0.18 / 76.18 ms │     no change │
│ QQuery 18 │     101.69 / 103.04 ±1.05 / 104.84 ms │         102.11 / 103.88 ±3.28 / 110.44 ms │     no change │
│ QQuery 19 │        40.51 / 40.79 ±0.18 / 41.06 ms │            40.70 / 41.10 ±0.27 / 41.49 ms │     no change │
│ QQuery 20 │        30.03 / 30.43 ±0.31 / 30.94 ms │            30.69 / 30.95 ±0.21 / 31.30 ms │     no change │
│ QQuery 21 │        16.19 / 16.47 ±0.20 / 16.71 ms │            16.27 / 16.53 ±0.24 / 16.93 ms │     no change │
│ QQuery 22 │        67.70 / 70.02 ±1.79 / 72.64 ms │            68.16 / 69.46 ±1.77 / 72.92 ms │     no change │
│ QQuery 23 │     283.88 / 286.16 ±1.73 / 288.02 ms │         288.15 / 291.58 ±4.24 / 299.85 ms │     no change │
│ QQuery 24 │     183.58 / 187.22 ±3.15 / 192.03 ms │         187.38 / 191.43 ±4.61 / 200.07 ms │     no change │
│ QQuery 25 │     102.03 / 105.09 ±3.79 / 112.50 ms │         103.04 / 105.90 ±2.19 / 109.67 ms │     no change │
│ QQuery 26 │        43.11 / 43.58 ±0.39 / 44.19 ms │            43.44 / 43.76 ±0.32 / 44.35 ms │     no change │
│ QQuery 27 │           5.17 / 5.29 ±0.10 / 5.47 ms │               5.09 / 5.33 ±0.17 / 5.56 ms │     no change │
│ QQuery 28 │        57.27 / 60.05 ±1.48 / 61.46 ms │            61.28 / 61.61 ±0.28 / 61.98 ms │     no change │
│ QQuery 29 │       91.31 / 95.58 ±6.75 / 109.04 ms │           91.07 / 96.64 ±7.18 / 110.53 ms │     no change │
│ QQuery 30 │        35.16 / 35.76 ±0.64 / 36.94 ms │            35.07 / 35.75 ±0.41 / 36.25 ms │     no change │
│ QQuery 31 │     106.96 / 107.29 ±0.38 / 108.01 ms │         107.32 / 108.34 ±1.03 / 110.29 ms │     no change │
│ QQuery 32 │        18.98 / 19.61 ±0.37 / 20.13 ms │            19.13 / 21.44 ±3.79 / 28.99 ms │  1.09x slower │
│ QQuery 33 │        35.63 / 36.38 ±0.47 / 37.08 ms │            36.14 / 36.99 ±0.89 / 38.69 ms │     no change │
│ QQuery 34 │           5.38 / 5.54 ±0.20 / 5.94 ms │               5.51 / 5.67 ±0.19 / 6.05 ms │     no change │
│ QQuery 35 │        68.18 / 68.39 ±0.20 / 68.69 ms │            68.14 / 68.57 ±0.44 / 69.38 ms │     no change │
│ QQuery 36 │           5.67 / 5.79 ±0.15 / 6.07 ms │               5.73 / 5.93 ±0.24 / 6.39 ms │     no change │
│ QQuery 37 │           7.22 / 7.28 ±0.05 / 7.36 ms │               7.22 / 7.39 ±0.14 / 7.64 ms │     no change │
│ QQuery 38 │        62.74 / 63.02 ±0.20 / 63.31 ms │            62.90 / 63.12 ±0.20 / 63.49 ms │     no change │
│ QQuery 39 │        70.66 / 72.01 ±1.40 / 74.66 ms │            70.30 / 72.86 ±3.43 / 79.63 ms │     no change │
│ QQuery 40 │        21.96 / 22.50 ±0.40 / 23.16 ms │            22.32 / 22.71 ±0.28 / 23.12 ms │     no change │
│ QQuery 41 │        14.10 / 14.17 ±0.08 / 14.32 ms │            14.00 / 14.12 ±0.10 / 14.31 ms │     no change │
│ QQuery 42 │        23.10 / 23.28 ±0.13 / 23.44 ms │            23.20 / 23.93 ±0.50 / 24.50 ms │     no change │
│ QQuery 43 │           4.53 / 4.66 ±0.15 / 4.96 ms │               4.51 / 4.60 ±0.15 / 4.89 ms │     no change │
│ QQuery 44 │           5.46 / 5.53 ±0.08 / 5.67 ms │               5.22 / 5.34 ±0.09 / 5.50 ms │     no change │
│ QQuery 45 │        34.32 / 34.65 ±0.34 / 35.25 ms │            34.57 / 34.95 ±0.43 / 35.77 ms │     no change │
│ QQuery 46 │        10.89 / 11.11 ±0.19 / 11.37 ms │            10.80 / 10.98 ±0.17 / 11.23 ms │     no change │
│ QQuery 47 │     197.79 / 200.20 ±1.73 / 202.20 ms │         197.76 / 199.72 ±1.32 / 201.82 ms │     no change │
│ QQuery 48 │        89.21 / 90.09 ±0.62 / 90.84 ms │            88.98 / 91.16 ±2.94 / 96.96 ms │     no change │
│ QQuery 49 │        68.09 / 68.71 ±0.58 / 69.70 ms │            68.50 / 69.27 ±0.93 / 71.02 ms │     no change │
│ QQuery 50 │        55.80 / 56.11 ±0.25 / 56.54 ms │            56.09 / 56.80 ±1.03 / 58.83 ms │     no change │
│ QQuery 51 │       92.25 / 94.85 ±3.14 / 100.77 ms │           93.05 / 95.18 ±3.00 / 101.10 ms │     no change │
│ QQuery 52 │        23.15 / 23.58 ±0.34 / 24.18 ms │            23.68 / 23.96 ±0.28 / 24.42 ms │     no change │
│ QQuery 53 │        28.60 / 28.78 ±0.11 / 28.92 ms │            28.57 / 28.91 ±0.19 / 29.09 ms │     no change │
│ QQuery 54 │        51.66 / 51.93 ±0.31 / 52.47 ms │            52.23 / 52.42 ±0.19 / 52.66 ms │     no change │
│ QQuery 55 │        22.83 / 23.04 ±0.17 / 23.33 ms │            22.98 / 24.93 ±2.35 / 28.91 ms │  1.08x slower │
│ QQuery 56 │        36.99 / 40.01 ±3.15 / 45.12 ms │            37.37 / 37.86 ±0.37 / 38.29 ms │ +1.06x faster │
│ QQuery 57 │     149.45 / 150.79 ±0.89 / 151.77 ms │         149.14 / 151.41 ±2.04 / 154.74 ms │     no change │
│ QQuery 58 │     100.87 / 103.64 ±1.84 / 106.56 ms │         103.04 / 103.64 ±0.78 / 105.16 ms │     no change │
│ QQuery 59 │     105.17 / 106.51 ±1.26 / 108.79 ms │         106.02 / 107.66 ±1.97 / 111.46 ms │     no change │
│ QQuery 60 │        37.65 / 38.30 ±0.49 / 38.93 ms │            37.79 / 38.29 ±0.60 / 39.46 ms │     no change │
│ QQuery 61 │           8.86 / 9.02 ±0.24 / 9.50 ms │               8.88 / 9.03 ±0.21 / 9.44 ms │     no change │
│ QQuery 62 │        36.87 / 37.12 ±0.28 / 37.64 ms │            36.94 / 37.29 ±0.28 / 37.79 ms │     no change │
│ QQuery 63 │        28.64 / 28.94 ±0.17 / 29.11 ms │            28.59 / 28.80 ±0.15 / 29.01 ms │     no change │
│ QQuery 64 │     345.17 / 348.31 ±2.31 / 351.57 ms │         348.47 / 354.53 ±4.52 / 362.50 ms │     no change │
│ QQuery 65 │        74.15 / 74.73 ±0.70 / 76.03 ms │            74.63 / 75.49 ±0.75 / 76.55 ms │     no change │
│ QQuery 66 │        70.18 / 70.54 ±0.37 / 71.16 ms │            70.89 / 71.52 ±0.49 / 72.20 ms │     no change │
│ QQuery 67 │     213.81 / 217.22 ±3.83 / 224.02 ms │         216.06 / 219.49 ±2.27 / 221.92 ms │     no change │
│ QQuery 68 │           6.52 / 6.66 ±0.17 / 6.99 ms │               6.38 / 6.63 ±0.30 / 7.17 ms │     no change │
│ QQuery 69 │        52.85 / 53.38 ±0.48 / 54.03 ms │            52.87 / 53.09 ±0.19 / 53.40 ms │     no change │
│ QQuery 70 │        95.19 / 95.82 ±0.44 / 96.32 ms │            94.92 / 95.62 ±0.84 / 97.28 ms │     no change │
│ QQuery 71 │        33.99 / 36.97 ±5.11 / 47.18 ms │            34.06 / 36.29 ±3.43 / 43.10 ms │     no change │
│ QQuery 72 │ 1609.58 / 1679.67 ±44.17 / 1745.29 ms │     1479.13 / 1526.78 ±29.69 / 1562.12 ms │ +1.10x faster │
│ QQuery 73 │           5.30 / 5.47 ±0.25 / 5.98 ms │               5.14 / 5.56 ±0.50 / 6.48 ms │     no change │
│ QQuery 74 │     242.69 / 248.78 ±4.86 / 255.54 ms │         239.08 / 242.55 ±3.19 / 246.53 ms │     no change │
│ QQuery 75 │     138.15 / 139.87 ±2.43 / 144.66 ms │         135.95 / 137.14 ±1.51 / 140.06 ms │     no change │
│ QQuery 76 │        33.85 / 34.33 ±0.26 / 34.53 ms │            33.33 / 33.59 ±0.26 / 33.96 ms │     no change │
│ QQuery 77 │        58.80 / 59.17 ±0.30 / 59.49 ms │            58.20 / 59.51 ±1.12 / 61.35 ms │     no change │
│ QQuery 78 │    255.36 / 265.78 ±11.45 / 286.21 ms │         252.07 / 261.54 ±7.47 / 274.17 ms │     no change │
│ QQuery 79 │        67.04 / 67.44 ±0.27 / 67.90 ms │            66.51 / 66.80 ±0.19 / 67.03 ms │     no change │
│ QQuery 80 │        92.54 / 93.74 ±1.16 / 95.78 ms │            92.12 / 94.18 ±1.44 / 96.34 ms │     no change │
│ QQuery 81 │        28.38 / 32.00 ±4.47 / 39.29 ms │            27.26 / 28.39 ±1.76 / 31.87 ms │ +1.13x faster │
│ QQuery 82 │        17.06 / 17.64 ±0.52 / 18.37 ms │            16.61 / 16.81 ±0.15 / 17.05 ms │     no change │
│ QQuery 83 │        31.70 / 32.19 ±0.36 / 32.62 ms │            31.09 / 31.27 ±0.18 / 31.58 ms │     no change │
│ QQuery 84 │        29.96 / 30.22 ±0.26 / 30.68 ms │            29.70 / 29.86 ±0.12 / 30.03 ms │     no change │
│ QQuery 85 │        94.02 / 94.33 ±0.22 / 94.65 ms │           92.64 / 96.38 ±3.23 / 102.40 ms │     no change │
│ QQuery 86 │        23.24 / 26.16 ±2.98 / 31.22 ms │            23.37 / 23.52 ±0.20 / 23.92 ms │ +1.11x faster │
│ QQuery 87 │        64.02 / 64.83 ±0.82 / 66.16 ms │            62.83 / 63.71 ±0.48 / 64.19 ms │     no change │
│ QQuery 88 │        59.44 / 59.95 ±0.39 / 60.42 ms │            59.12 / 60.60 ±2.50 / 65.58 ms │     no change │
│ QQuery 89 │        35.30 / 35.71 ±0.52 / 36.72 ms │            35.25 / 36.40 ±0.82 / 37.56 ms │     no change │
│ QQuery 90 │        15.80 / 15.97 ±0.21 / 16.36 ms │            15.67 / 15.89 ±0.16 / 16.18 ms │     no change │
│ QQuery 91 │        39.86 / 41.26 ±2.32 / 45.88 ms │            38.70 / 38.88 ±0.16 / 39.15 ms │ +1.06x faster │
│ QQuery 92 │        27.94 / 28.45 ±0.37 / 28.97 ms │            27.69 / 28.30 ±0.41 / 28.86 ms │     no change │
│ QQuery 93 │        48.57 / 49.51 ±0.80 / 50.52 ms │            47.77 / 48.55 ±0.62 / 49.27 ms │     no change │
│ QQuery 94 │        36.75 / 37.22 ±0.34 / 37.75 ms │            36.89 / 39.18 ±2.36 / 43.38 ms │  1.05x slower │
│ QQuery 95 │        77.39 / 79.50 ±2.62 / 84.24 ms │            79.09 / 79.98 ±0.79 / 81.21 ms │     no change │
│ QQuery 96 │        23.22 / 23.46 ±0.25 / 23.94 ms │            23.16 / 23.34 ±0.18 / 23.62 ms │     no change │
│ QQuery 97 │        49.81 / 50.41 ±0.65 / 51.68 ms │            49.02 / 50.09 ±1.18 / 52.37 ms │     no change │
│ QQuery 98 │        37.02 / 37.28 ±0.27 / 37.76 ms │            37.20 / 37.31 ±0.09 / 37.41 ms │     no change │
│ QQuery 99 │        55.19 / 55.58 ±0.35 / 56.18 ms │            55.22 / 56.74 ±2.40 / 61.53 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9018.21ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 8895.18ms │
│ Average Time (HEAD)                                      │   91.09ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   89.85ms │
│ Queries Faster                                           │         5 │
│ Queries Slower                                           │         3 │
│ Queries with No Change                                   │        91 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.1 GiB
Avg memory 1.3 GiB
CPU user 184.5s
CPU sys 5.1s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 45.0s
Peak memory 2.1 GiB
Avg memory 1.5 GiB
CPU user 178.5s
CPU sys 5.3s
Peak spill 0 B

File an issue against this benchmark runner

@Rich-T-kid
Rich-T-kid marked this pull request as ready for review November 10, 2026 04:31
@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

@kumarUjjawal could you take a look

@kumarUjjawal kumarUjjawal left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you @Rich-T-kid

Left few comments please take a look

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

run benchmark dictionary_group_values tpch tpcds clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

@Rich-T-kid

Copy link
Copy Markdown
Contributor Author

@kumarUjjawal PTAL

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6827465841-4826-bhp7v 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6827465841-4827-stp6b 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6827465841-4825-h8z4l 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6827465841-4824-v5prm 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark dictionary_group_values
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 40.96 ms │                                  40.72 ms │ no change │
│ QQuery 2  │ 19.92 ms │                                  19.97 ms │ no change │
│ QQuery 3  │ 29.65 ms │                                  29.89 ms │ no change │
│ QQuery 4  │ 17.85 ms │                                  17.90 ms │ no change │
│ QQuery 5  │ 37.40 ms │                                  38.16 ms │ no change │
│ QQuery 6  │ 14.38 ms │                                  14.38 ms │ no change │
│ QQuery 7  │ 46.67 ms │                                  44.97 ms │ no change │
│ QQuery 8  │ 42.21 ms │                                  42.64 ms │ no change │
│ QQuery 9  │ 51.99 ms │                                  53.14 ms │ no change │
│ QQuery 10 │ 61.04 ms │                                  62.11 ms │ no change │
│ QQuery 11 │ 14.48 ms │                                  14.62 ms │ no change │
│ QQuery 12 │ 21.15 ms │                                  21.67 ms │ no change │
│ QQuery 13 │ 50.93 ms │                                  51.31 ms │ no change │
│ QQuery 14 │ 25.05 ms │                                  25.27 ms │ no change │
│ QQuery 15 │ 32.46 ms │                                  32.90 ms │ no change │
│ QQuery 16 │ 14.06 ms │                                  14.26 ms │ no change │
│ QQuery 17 │ 77.62 ms │                                  77.92 ms │ no change │
│ QQuery 18 │ 67.11 ms │                                  68.33 ms │ no change │
│ QQuery 19 │ 30.33 ms │                                  30.64 ms │ no change │
│ QQuery 20 │ 34.72 ms │                                  36.24 ms │ no change │
│ QQuery 21 │ 57.17 ms │                                  58.44 ms │ no change │
│ QQuery 22 │ 16.66 ms │                                  16.64 ms │ no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 803.81ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 812.10ms │
│ Average Time (HEAD)                                      │  36.54ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  36.91ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ rich-t-kid_single-dictionary-group-values ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 40.96 / 42.60 ±1.43 / 44.93 ms │            40.72 / 42.12 ±1.45 / 44.18 ms │ no change │
│ QQuery 2  │ 19.92 / 20.27 ±0.32 / 20.86 ms │            19.97 / 20.11 ±0.12 / 20.26 ms │ no change │
│ QQuery 3  │ 29.65 / 30.17 ±0.30 / 30.52 ms │            29.89 / 30.13 ±0.35 / 30.81 ms │ no change │
│ QQuery 4  │ 17.85 / 18.36 ±0.59 / 19.44 ms │            17.90 / 18.01 ±0.09 / 18.15 ms │ no change │
│ QQuery 5  │ 37.40 / 37.78 ±0.37 / 38.34 ms │            38.16 / 38.33 ±0.10 / 38.44 ms │ no change │
│ QQuery 6  │ 14.38 / 14.49 ±0.07 / 14.58 ms │            14.38 / 14.73 ±0.20 / 14.94 ms │ no change │
│ QQuery 7  │ 46.67 / 47.49 ±0.59 / 48.36 ms │            44.97 / 46.82 ±1.78 / 49.18 ms │ no change │
│ QQuery 8  │ 42.21 / 42.70 ±0.33 / 43.27 ms │            42.64 / 43.49 ±1.13 / 45.71 ms │ no change │
│ QQuery 9  │ 51.99 / 52.37 ±0.34 / 52.95 ms │            53.14 / 54.96 ±1.41 / 56.82 ms │ no change │
│ QQuery 10 │ 61.04 / 62.28 ±1.26 / 64.65 ms │            62.11 / 64.28 ±1.93 / 67.74 ms │ no change │
│ QQuery 11 │ 14.48 / 14.62 ±0.11 / 14.75 ms │            14.62 / 14.71 ±0.08 / 14.86 ms │ no change │
│ QQuery 12 │ 21.15 / 22.19 ±1.64 / 25.45 ms │            21.67 / 23.01 ±2.13 / 27.26 ms │ no change │
│ QQuery 13 │ 50.93 / 51.30 ±0.25 / 51.68 ms │            51.31 / 52.04 ±0.75 / 53.47 ms │ no change │
│ QQuery 14 │ 25.05 / 25.22 ±0.10 / 25.33 ms │            25.27 / 25.54 ±0.23 / 25.95 ms │ no change │
│ QQuery 15 │ 32.46 / 32.69 ±0.12 / 32.81 ms │            32.90 / 33.11 ±0.22 / 33.50 ms │ no change │
│ QQuery 16 │ 14.06 / 14.16 ±0.08 / 14.26 ms │            14.26 / 14.40 ±0.13 / 14.64 ms │ no change │
│ QQuery 17 │ 77.62 / 78.80 ±1.45 / 81.64 ms │            77.92 / 80.24 ±2.49 / 85.05 ms │ no change │
│ QQuery 18 │ 67.11 / 71.19 ±2.20 / 73.75 ms │            68.33 / 72.17 ±2.60 / 75.14 ms │ no change │
│ QQuery 19 │ 30.33 / 30.90 ±0.41 / 31.37 ms │            30.64 / 31.37 ±0.41 / 31.73 ms │ no change │
│ QQuery 20 │ 34.72 / 35.46 ±0.54 / 36.36 ms │            36.24 / 36.54 ±0.35 / 37.23 ms │ no change │
│ QQuery 21 │ 57.17 / 58.66 ±2.48 / 63.60 ms │            58.44 / 60.69 ±2.70 / 65.99 ms │ no change │
│ QQuery 22 │ 16.66 / 17.37 ±0.51 / 17.94 ms │            16.64 / 16.96 ±0.29 / 17.50 ms │ no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 821.08ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 833.75ms │
│ Average Time (HEAD)                                      │  37.32ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │  37.90ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 636.8 MiB
CPU user 21.3s
CPU sys 1.8s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 636.9 MiB
CPU user 21.0s
CPU sys 1.8s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.54 ms │                                   5.75 ms │    no change │
│ QQuery 2  │   76.84 ms │                                  77.06 ms │    no change │
│ QQuery 3  │   29.40 ms │                                  29.38 ms │    no change │
│ QQuery 4  │  613.47 ms │                                 622.48 ms │    no change │
│ QQuery 5  │   51.85 ms │                                  50.62 ms │    no change │
│ QQuery 6  │   35.43 ms │                                  34.22 ms │    no change │
│ QQuery 7  │   72.81 ms │                                  72.04 ms │    no change │
│ QQuery 8  │   40.12 ms │                                  39.48 ms │    no change │
│ QQuery 9  │   54.20 ms │                                  53.70 ms │    no change │
│ QQuery 10 │   60.76 ms │                                  60.47 ms │    no change │
│ QQuery 11 │  417.22 ms │                                 425.69 ms │    no change │
│ QQuery 12 │   27.60 ms │                                  26.80 ms │    no change │
│ QQuery 13 │  115.26 ms │                                 113.61 ms │    no change │
│ QQuery 14 │  388.01 ms │                                 381.30 ms │    no change │
│ QQuery 15 │   54.44 ms │                                  53.49 ms │    no change │
│ QQuery 16 │    6.39 ms │                                   6.40 ms │    no change │
│ QQuery 17 │   79.05 ms │                                  79.03 ms │    no change │
│ QQuery 18 │  107.60 ms │                                 105.11 ms │    no change │
│ QQuery 19 │   42.75 ms │                                  42.08 ms │    no change │
│ QQuery 20 │   32.27 ms │                                  31.98 ms │    no change │
│ QQuery 21 │   17.78 ms │                                  17.57 ms │    no change │
│ QQuery 22 │   72.05 ms │                                  72.10 ms │    no change │
│ QQuery 23 │  315.77 ms │                                 304.58 ms │    no change │
│ QQuery 24 │  204.55 ms │                                 200.48 ms │    no change │
│ QQuery 25 │  106.03 ms │                                 106.35 ms │    no change │
│ QQuery 26 │   45.77 ms │                                  45.02 ms │    no change │
│ QQuery 27 │    5.83 ms │                                   5.58 ms │    no change │
│ QQuery 28 │   58.69 ms │                                  61.81 ms │ 1.05x slower │
│ QQuery 29 │   95.93 ms │                                  94.30 ms │    no change │
│ QQuery 30 │   37.43 ms │                                  37.69 ms │    no change │
│ QQuery 31 │  111.22 ms │                                 110.80 ms │    no change │
│ QQuery 32 │   20.39 ms │                                  20.15 ms │    no change │
│ QQuery 33 │   36.71 ms │                                  37.27 ms │    no change │
│ QQuery 34 │    5.65 ms │                                   5.77 ms │    no change │
│ QQuery 35 │   70.24 ms │                                  71.36 ms │    no change │
│ QQuery 36 │    5.77 ms │                                   6.18 ms │ 1.07x slower │
│ QQuery 37 │    7.30 ms │                                   7.91 ms │ 1.08x slower │
│ QQuery 38 │   65.13 ms │                                  65.84 ms │    no change │
│ QQuery 39 │   72.31 ms │                                  73.37 ms │    no change │
│ QQuery 40 │   22.57 ms │                                  23.39 ms │    no change │
│ QQuery 41 │   14.23 ms │                                  14.48 ms │    no change │
│ QQuery 42 │   23.57 ms │                                  24.25 ms │    no change │
│ QQuery 43 │    4.62 ms │                                   4.93 ms │ 1.07x slower │
│ QQuery 44 │    5.37 ms │                                   5.84 ms │ 1.09x slower │
│ QQuery 45 │   36.14 ms │                                  36.62 ms │    no change │
│ QQuery 46 │   11.35 ms │                                  11.97 ms │ 1.05x slower │
│ QQuery 47 │  205.28 ms │                                 205.98 ms │    no change │
│ QQuery 48 │   91.16 ms │                                  91.35 ms │    no change │
│ QQuery 49 │   69.08 ms │                                  69.81 ms │    no change │
│ QQuery 50 │   56.24 ms │                                  58.14 ms │    no change │
│ QQuery 51 │   93.43 ms │                                  93.81 ms │    no change │
│ QQuery 52 │   24.61 ms │                                  24.98 ms │    no change │
│ QQuery 53 │   29.86 ms │                                  29.93 ms │    no change │
│ QQuery 54 │   55.09 ms │                                  53.67 ms │    no change │
│ QQuery 55 │   24.44 ms │                                  24.19 ms │    no change │
│ QQuery 56 │   38.78 ms │                                  38.62 ms │    no change │
│ QQuery 57 │  155.50 ms │                                 152.66 ms │    no change │
│ QQuery 58 │  105.69 ms │                                 103.83 ms │    no change │
│ QQuery 59 │  107.73 ms │                                 106.57 ms │    no change │
│ QQuery 60 │   38.99 ms │                                  38.80 ms │    no change │
│ QQuery 61 │    9.29 ms │                                   9.09 ms │    no change │
│ QQuery 62 │   38.60 ms │                                  38.38 ms │    no change │
│ QQuery 63 │   29.87 ms │                                  30.01 ms │    no change │
│ QQuery 64 │  359.32 ms │                                 362.75 ms │    no change │
│ QQuery 65 │   79.25 ms │                                  79.10 ms │    no change │
│ QQuery 66 │   74.21 ms │                                  74.24 ms │    no change │
│ QQuery 67 │  242.01 ms │                                 246.85 ms │    no change │
│ QQuery 68 │    6.81 ms │                                   7.10 ms │    no change │
│ QQuery 69 │   54.63 ms │                                  55.08 ms │    no change │
│ QQuery 70 │   97.05 ms │                                  97.18 ms │    no change │
│ QQuery 71 │   35.44 ms │                                  36.28 ms │    no change │
│ QQuery 72 │ 1605.34 ms │                                1649.11 ms │    no change │
│ QQuery 73 │    5.22 ms │                                   5.82 ms │ 1.11x slower │
│ QQuery 74 │  254.51 ms │                                 275.65 ms │ 1.08x slower │
│ QQuery 75 │  141.40 ms │                                 138.48 ms │    no change │
│ QQuery 76 │   35.66 ms │                                  34.71 ms │    no change │
│ QQuery 77 │   60.35 ms │                                  60.99 ms │    no change │
│ QQuery 78 │  277.15 ms │                                 266.91 ms │    no change │
│ QQuery 79 │   68.57 ms │                                  68.10 ms │    no change │
│ QQuery 80 │   95.78 ms │                                  95.54 ms │    no change │
│ QQuery 81 │   29.12 ms │                                  28.33 ms │    no change │
│ QQuery 82 │   17.77 ms │                                  17.57 ms │    no change │
│ QQuery 83 │   33.75 ms │                                  32.89 ms │    no change │
│ QQuery 84 │   31.83 ms │                                  30.90 ms │    no change │
│ QQuery 85 │   97.76 ms │                                  94.75 ms │    no change │
│ QQuery 86 │   24.98 ms │                                  24.29 ms │    no change │
│ QQuery 87 │   66.83 ms │                                  66.83 ms │    no change │
│ QQuery 88 │   60.98 ms │                                  60.14 ms │    no change │
│ QQuery 89 │   36.33 ms │                                  36.15 ms │    no change │
│ QQuery 90 │   16.52 ms │                                  16.47 ms │    no change │
│ QQuery 91 │   39.52 ms │                                  40.14 ms │    no change │
│ QQuery 92 │   29.77 ms │                                  30.06 ms │    no change │
│ QQuery 93 │   48.15 ms │                                  49.84 ms │    no change │
│ QQuery 94 │   38.86 ms │                                  39.09 ms │    no change │
│ QQuery 95 │   79.66 ms │                                  80.96 ms │    no change │
│ QQuery 96 │   24.16 ms │                                  23.91 ms │    no change │
│ QQuery 97 │   50.78 ms │                                  50.52 ms │    no change │
│ QQuery 98 │   38.81 ms │                                  38.72 ms │    no change │
│ QQuery 99 │   56.93 ms │                                  56.74 ms │    no change │
└───────────┴────────────┴───────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9272.24ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 9318.28ms │
│ Average Time (HEAD)                                      │   93.66ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   94.12ms │
│ Queries Faster                                           │         0 │
│ Queries Slower                                           │         8 │
│ Queries with No Change                                   │        91 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.54 / 6.22 ±1.10 / 8.41 ms │               5.75 / 6.30 ±1.00 / 8.31 ms │     no change │
│ QQuery 2  │        76.84 / 77.18 ±0.25 / 77.59 ms │            77.06 / 77.34 ±0.21 / 77.71 ms │     no change │
│ QQuery 3  │        29.40 / 29.60 ±0.16 / 29.89 ms │            29.38 / 29.64 ±0.21 / 30.01 ms │     no change │
│ QQuery 4  │     613.47 / 624.13 ±7.25 / 631.87 ms │         622.48 / 632.22 ±7.44 / 643.20 ms │     no change │
│ QQuery 5  │        51.85 / 52.06 ±0.19 / 52.34 ms │            50.62 / 51.08 ±0.25 / 51.27 ms │     no change │
│ QQuery 6  │        35.43 / 36.35 ±1.50 / 39.34 ms │            34.22 / 35.76 ±1.50 / 38.61 ms │     no change │
│ QQuery 7  │        72.81 / 73.97 ±0.89 / 75.51 ms │            72.04 / 72.97 ±0.59 / 73.72 ms │     no change │
│ QQuery 8  │        40.12 / 40.43 ±0.16 / 40.57 ms │            39.48 / 40.01 ±0.38 / 40.57 ms │     no change │
│ QQuery 9  │        54.20 / 55.65 ±1.24 / 57.54 ms │            53.70 / 55.15 ±0.90 / 56.49 ms │     no change │
│ QQuery 10 │        60.76 / 62.66 ±2.89 / 68.42 ms │            60.47 / 63.07 ±3.93 / 70.82 ms │     no change │
│ QQuery 11 │     417.22 / 419.64 ±2.16 / 422.27 ms │         425.69 / 428.40 ±2.53 / 432.95 ms │     no change │
│ QQuery 12 │        27.60 / 27.80 ±0.20 / 28.18 ms │            26.80 / 27.16 ±0.26 / 27.57 ms │     no change │
│ QQuery 13 │     115.26 / 117.06 ±2.25 / 121.46 ms │         113.61 / 113.99 ±0.24 / 114.33 ms │     no change │
│ QQuery 14 │     388.01 / 390.73 ±2.19 / 394.16 ms │         381.30 / 384.89 ±4.06 / 392.26 ms │     no change │
│ QQuery 15 │        54.44 / 55.58 ±1.09 / 56.92 ms │            53.49 / 53.78 ±0.22 / 54.13 ms │     no change │
│ QQuery 16 │           6.39 / 6.64 ±0.14 / 6.81 ms │              6.40 / 7.94 ±2.85 / 13.64 ms │  1.20x slower │
│ QQuery 17 │        79.05 / 80.47 ±1.88 / 84.12 ms │            79.03 / 80.08 ±0.68 / 81.09 ms │     no change │
│ QQuery 18 │     107.60 / 109.49 ±2.45 / 114.20 ms │         105.11 / 107.30 ±1.84 / 110.60 ms │     no change │
│ QQuery 19 │        42.75 / 43.26 ±0.38 / 43.82 ms │            42.08 / 42.44 ±0.47 / 43.33 ms │     no change │
│ QQuery 20 │        32.27 / 32.61 ±0.32 / 33.12 ms │            31.98 / 32.50 ±0.44 / 33.30 ms │     no change │
│ QQuery 21 │        17.78 / 17.83 ±0.05 / 17.90 ms │            17.57 / 18.07 ±0.81 / 19.68 ms │     no change │
│ QQuery 22 │        72.05 / 73.08 ±0.66 / 74.12 ms │            72.10 / 72.79 ±0.66 / 73.90 ms │     no change │
│ QQuery 23 │     315.77 / 319.32 ±4.02 / 326.76 ms │         304.58 / 310.64 ±5.95 / 320.83 ms │     no change │
│ QQuery 24 │     204.55 / 206.78 ±1.77 / 209.80 ms │         200.48 / 203.87 ±3.58 / 210.20 ms │     no change │
│ QQuery 25 │     106.03 / 109.10 ±3.08 / 114.91 ms │         106.35 / 107.56 ±1.15 / 109.47 ms │     no change │
│ QQuery 26 │        45.77 / 47.78 ±3.28 / 54.33 ms │            45.02 / 47.06 ±3.15 / 53.30 ms │     no change │
│ QQuery 27 │           5.83 / 6.00 ±0.13 / 6.24 ms │               5.58 / 5.75 ±0.14 / 5.95 ms │     no change │
│ QQuery 28 │        58.69 / 62.06 ±1.69 / 63.12 ms │            61.81 / 62.30 ±0.73 / 63.75 ms │     no change │
│ QQuery 29 │       95.93 / 98.18 ±2.97 / 104.00 ms │            94.30 / 95.51 ±1.04 / 96.71 ms │     no change │
│ QQuery 30 │        37.43 / 39.47 ±3.62 / 46.70 ms │            37.69 / 39.30 ±2.04 / 43.29 ms │     no change │
│ QQuery 31 │     111.22 / 112.04 ±0.64 / 113.16 ms │         110.80 / 111.60 ±0.60 / 112.50 ms │     no change │
│ QQuery 32 │        20.39 / 20.84 ±0.43 / 21.42 ms │            20.15 / 20.54 ±0.22 / 20.77 ms │     no change │
│ QQuery 33 │        36.71 / 37.28 ±0.44 / 37.79 ms │            37.27 / 37.40 ±0.10 / 37.55 ms │     no change │
│ QQuery 34 │           5.65 / 5.90 ±0.23 / 6.31 ms │               5.77 / 5.91 ±0.15 / 6.19 ms │     no change │
│ QQuery 35 │        70.24 / 71.99 ±1.11 / 73.49 ms │            71.36 / 73.53 ±2.48 / 78.23 ms │     no change │
│ QQuery 36 │           5.77 / 5.97 ±0.25 / 6.46 ms │               6.18 / 6.36 ±0.18 / 6.68 ms │  1.07x slower │
│ QQuery 37 │           7.30 / 7.54 ±0.16 / 7.73 ms │               7.91 / 8.10 ±0.17 / 8.38 ms │  1.07x slower │
│ QQuery 38 │        65.13 / 65.94 ±0.56 / 66.66 ms │            65.84 / 66.25 ±0.26 / 66.54 ms │     no change │
│ QQuery 39 │        72.31 / 73.57 ±2.20 / 77.95 ms │            73.37 / 75.72 ±2.40 / 80.30 ms │     no change │
│ QQuery 40 │        22.57 / 22.99 ±0.46 / 23.78 ms │            23.39 / 24.00 ±0.62 / 25.08 ms │     no change │
│ QQuery 41 │        14.23 / 14.43 ±0.20 / 14.79 ms │            14.48 / 14.72 ±0.16 / 14.94 ms │     no change │
│ QQuery 42 │        23.57 / 24.10 ±0.41 / 24.66 ms │            24.25 / 24.69 ±0.53 / 25.74 ms │     no change │
│ QQuery 43 │           4.62 / 4.73 ±0.14 / 5.01 ms │               4.93 / 5.09 ±0.11 / 5.27 ms │  1.08x slower │
│ QQuery 44 │           5.37 / 5.50 ±0.10 / 5.68 ms │               5.84 / 5.96 ±0.08 / 6.09 ms │  1.08x slower │
│ QQuery 45 │        36.14 / 37.43 ±1.51 / 40.31 ms │            36.62 / 38.35 ±1.12 / 40.01 ms │     no change │
│ QQuery 46 │        11.35 / 11.76 ±0.38 / 12.41 ms │            11.97 / 12.37 ±0.45 / 13.18 ms │  1.05x slower │
│ QQuery 47 │     205.28 / 211.12 ±3.51 / 215.61 ms │         205.98 / 212.73 ±3.81 / 217.42 ms │     no change │
│ QQuery 48 │        91.16 / 91.91 ±0.50 / 92.73 ms │            91.35 / 92.72 ±1.00 / 94.47 ms │     no change │
│ QQuery 49 │        69.08 / 70.61 ±1.01 / 72.28 ms │            69.81 / 70.24 ±0.43 / 70.97 ms │     no change │
│ QQuery 50 │        56.24 / 57.23 ±0.76 / 58.35 ms │            58.14 / 58.29 ±0.15 / 58.57 ms │     no change │
│ QQuery 51 │       93.43 / 97.12 ±2.32 / 100.48 ms │           93.81 / 97.40 ±4.71 / 106.46 ms │     no change │
│ QQuery 52 │        24.61 / 24.91 ±0.18 / 25.09 ms │            24.98 / 25.16 ±0.12 / 25.32 ms │     no change │
│ QQuery 53 │        29.86 / 31.07 ±1.86 / 34.78 ms │            29.93 / 30.10 ±0.15 / 30.28 ms │     no change │
│ QQuery 54 │        55.09 / 55.33 ±0.17 / 55.59 ms │            53.67 / 54.29 ±0.47 / 54.83 ms │     no change │
│ QQuery 55 │        24.44 / 24.62 ±0.12 / 24.79 ms │            24.19 / 26.05 ±2.37 / 30.61 ms │  1.06x slower │
│ QQuery 56 │        38.78 / 39.64 ±0.50 / 40.26 ms │            38.62 / 39.42 ±0.65 / 40.23 ms │     no change │
│ QQuery 57 │     155.50 / 157.28 ±2.77 / 162.72 ms │         152.66 / 155.47 ±2.46 / 159.89 ms │     no change │
│ QQuery 58 │     105.69 / 107.10 ±1.14 / 108.34 ms │         103.83 / 105.24 ±0.88 / 106.21 ms │     no change │
│ QQuery 59 │     107.73 / 109.19 ±1.90 / 112.88 ms │         106.57 / 107.20 ±0.63 / 108.29 ms │     no change │
│ QQuery 60 │        38.99 / 39.75 ±0.47 / 40.45 ms │            38.80 / 39.53 ±0.88 / 40.98 ms │     no change │
│ QQuery 61 │          9.29 / 9.68 ±0.43 / 10.48 ms │               9.09 / 9.49 ±0.26 / 9.83 ms │     no change │
│ QQuery 62 │        38.60 / 38.84 ±0.27 / 39.35 ms │            38.38 / 38.94 ±0.35 / 39.35 ms │     no change │
│ QQuery 63 │        29.87 / 31.16 ±1.50 / 34.07 ms │            30.01 / 30.65 ±0.58 / 31.65 ms │     no change │
│ QQuery 64 │     359.32 / 364.11 ±4.34 / 369.82 ms │         362.75 / 368.77 ±4.60 / 374.95 ms │     no change │
│ QQuery 65 │        79.25 / 80.63 ±1.01 / 82.15 ms │            79.10 / 79.95 ±0.82 / 81.05 ms │     no change │
│ QQuery 66 │        74.21 / 76.47 ±3.53 / 83.49 ms │            74.24 / 76.63 ±2.51 / 81.32 ms │     no change │
│ QQuery 67 │     242.01 / 245.57 ±3.18 / 249.97 ms │         246.85 / 250.61 ±3.34 / 256.88 ms │     no change │
│ QQuery 68 │           6.81 / 7.01 ±0.23 / 7.46 ms │               7.10 / 7.21 ±0.09 / 7.36 ms │     no change │
│ QQuery 69 │        54.63 / 56.10 ±2.20 / 60.41 ms │            55.08 / 55.30 ±0.21 / 55.68 ms │     no change │
│ QQuery 70 │      97.05 / 101.30 ±4.83 / 110.73 ms │           97.18 / 99.23 ±1.92 / 102.79 ms │     no change │
│ QQuery 71 │        35.44 / 35.62 ±0.10 / 35.75 ms │            36.28 / 38.04 ±2.81 / 43.56 ms │  1.07x slower │
│ QQuery 72 │ 1605.34 / 1691.48 ±56.85 / 1769.37 ms │     1649.11 / 1751.43 ±55.36 / 1815.86 ms │     no change │
│ QQuery 73 │           5.22 / 5.41 ±0.23 / 5.86 ms │               5.82 / 5.98 ±0.09 / 6.09 ms │  1.11x slower │
│ QQuery 74 │    254.51 / 264.98 ±11.16 / 285.16 ms │         275.65 / 283.28 ±5.12 / 291.46 ms │  1.07x slower │
│ QQuery 75 │     141.40 / 143.77 ±1.63 / 145.85 ms │         138.48 / 142.03 ±3.92 / 149.32 ms │     no change │
│ QQuery 76 │        35.66 / 37.77 ±2.84 / 43.34 ms │            34.71 / 34.89 ±0.23 / 35.33 ms │ +1.08x faster │
│ QQuery 77 │        60.35 / 61.16 ±0.68 / 62.26 ms │            60.99 / 61.69 ±0.59 / 62.68 ms │     no change │
│ QQuery 78 │     277.15 / 283.19 ±8.49 / 299.92 ms │        266.91 / 284.87 ±14.33 / 310.25 ms │     no change │
│ QQuery 79 │        68.57 / 70.90 ±3.37 / 77.48 ms │            68.10 / 68.90 ±0.43 / 69.37 ms │     no change │
│ QQuery 80 │       95.78 / 98.54 ±2.87 / 103.68 ms │           95.54 / 97.41 ±1.78 / 100.21 ms │     no change │
│ QQuery 81 │        29.12 / 29.61 ±0.33 / 30.09 ms │            28.33 / 28.70 ±0.32 / 29.28 ms │     no change │
│ QQuery 82 │        17.77 / 18.20 ±0.56 / 19.31 ms │            17.57 / 17.81 ±0.16 / 18.07 ms │     no change │
│ QQuery 83 │        33.75 / 35.62 ±3.28 / 42.17 ms │            32.89 / 34.02 ±1.62 / 37.25 ms │     no change │
│ QQuery 84 │        31.83 / 32.46 ±0.49 / 33.18 ms │            30.90 / 31.32 ±0.47 / 32.23 ms │     no change │
│ QQuery 85 │       97.76 / 98.79 ±0.96 / 100.28 ms │            94.75 / 96.61 ±1.43 / 99.07 ms │     no change │
│ QQuery 86 │        24.98 / 25.26 ±0.19 / 25.51 ms │            24.29 / 24.64 ±0.49 / 25.59 ms │     no change │
│ QQuery 87 │        66.83 / 68.91 ±3.26 / 75.39 ms │            66.83 / 67.39 ±0.37 / 67.84 ms │     no change │
│ QQuery 88 │        60.98 / 61.24 ±0.20 / 61.55 ms │            60.14 / 60.85 ±0.50 / 61.48 ms │     no change │
│ QQuery 89 │        36.33 / 36.82 ±0.40 / 37.49 ms │            36.15 / 37.85 ±2.00 / 41.68 ms │     no change │
│ QQuery 90 │        16.52 / 16.87 ±0.42 / 17.70 ms │            16.47 / 16.69 ±0.17 / 16.92 ms │     no change │
│ QQuery 91 │        39.52 / 40.63 ±1.35 / 43.28 ms │            40.14 / 40.46 ±0.31 / 40.90 ms │     no change │
│ QQuery 92 │        29.77 / 30.40 ±0.56 / 31.10 ms │            30.06 / 30.44 ±0.30 / 30.80 ms │     no change │
│ QQuery 93 │        48.15 / 49.25 ±0.60 / 49.79 ms │            49.84 / 50.43 ±0.31 / 50.76 ms │     no change │
│ QQuery 94 │        38.86 / 39.66 ±1.20 / 42.04 ms │            39.09 / 39.98 ±1.08 / 42.08 ms │     no change │
│ QQuery 95 │        79.66 / 82.47 ±1.60 / 84.29 ms │            80.96 / 81.90 ±0.71 / 82.69 ms │     no change │
│ QQuery 96 │        24.16 / 24.34 ±0.23 / 24.80 ms │            23.91 / 24.08 ±0.16 / 24.35 ms │     no change │
│ QQuery 97 │        50.78 / 52.10 ±0.86 / 52.93 ms │            50.52 / 51.44 ±0.65 / 52.46 ms │     no change │
│ QQuery 98 │        38.81 / 39.72 ±1.21 / 42.04 ms │            38.72 / 39.10 ±0.30 / 39.50 ms │     no change │
│ QQuery 99 │        56.93 / 57.03 ±0.09 / 57.16 ms │            56.74 / 57.70 ±0.59 / 58.51 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9503.07ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 9566.02ms │
│ Average Time (HEAD)                                      │   95.99ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   96.63ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │        10 │
│ Queries with No Change                                   │        88 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 1.9 GiB
Avg memory 1.3 GiB
CPU user 193.6s
CPU sys 5.9s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.2 GiB
Avg memory 1.5 GiB
CPU user 198.2s
CPU sys 5.9s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.15 ms │                                   1.11 ms │     no change │
│ QQuery 1  │   11.72 ms │                                  11.54 ms │     no change │
│ QQuery 2  │   34.89 ms │                                  34.54 ms │     no change │
│ QQuery 3  │   30.65 ms │                                  30.28 ms │     no change │
│ QQuery 4  │  236.74 ms │                                 246.74 ms │     no change │
│ QQuery 5  │  833.84 ms │                                 738.85 ms │ +1.13x faster │
│ QQuery 6  │    1.26 ms │                                   1.15 ms │ +1.09x faster │
│ QQuery 7  │   13.09 ms │                                  12.84 ms │     no change │
│ QQuery 8  │  336.96 ms │                                 342.02 ms │     no change │
│ QQuery 9  │  481.34 ms │                                 489.34 ms │     no change │
│ QQuery 10 │  203.18 ms │                                 200.24 ms │     no change │
│ QQuery 11 │  214.23 ms │                                 212.57 ms │     no change │
│ QQuery 12 │  835.73 ms │                                 782.00 ms │ +1.07x faster │
│ QQuery 13 │ 1108.82 ms │                                1100.79 ms │     no change │
│ QQuery 14 │  867.00 ms │                                 876.54 ms │     no change │
│ QQuery 15 │  296.42 ms │                                 305.55 ms │     no change │
│ QQuery 16 │ 1345.32 ms │                                1378.29 ms │     no change │
│ QQuery 17 │ 1305.27 ms │                                1341.38 ms │     no change │
│ QQuery 18 │ 2101.48 ms │                                2107.64 ms │     no change │
│ QQuery 19 │   28.48 ms │                                  28.80 ms │     no change │
│ QQuery 20 │ 1063.35 ms │                                1056.67 ms │     no change │
│ QQuery 21 │ 1366.70 ms │                                1385.05 ms │     no change │
│ QQuery 22 │ 3270.97 ms │                                3297.05 ms │     no change │
│ QQuery 23 │ 9236.77 ms │                                9376.38 ms │     no change │
│ QQuery 24 │   97.23 ms │                                  96.21 ms │     no change │
│ QQuery 25 │  356.23 ms │                                 358.03 ms │     no change │
│ QQuery 26 │   98.45 ms │                                  97.28 ms │     no change │
│ QQuery 27 │ 1157.18 ms │                                1158.31 ms │     no change │
│ QQuery 28 │ 2572.88 ms │                                2519.63 ms │     no change │
│ QQuery 29 │   38.72 ms │                                  37.65 ms │     no change │
│ QQuery 30 │  557.93 ms │                                 558.05 ms │     no change │
│ QQuery 31 │  556.94 ms │                                 549.99 ms │     no change │
│ QQuery 32 │ 1090.10 ms │                                1088.35 ms │     no change │
│ QQuery 33 │ 3887.48 ms │                                3356.35 ms │ +1.16x faster │
│ QQuery 34 │ 3854.98 ms │                                3587.70 ms │ +1.07x faster │
│ QQuery 35 │  303.45 ms │                                 318.28 ms │     no change │
│ QQuery 36 │  135.90 ms │                                 123.67 ms │ +1.10x faster │
│ QQuery 37 │   97.46 ms │                                  95.05 ms │     no change │
│ QQuery 38 │   83.81 ms │                                  80.40 ms │     no change │
│ QQuery 39 │  306.14 ms │                                 314.79 ms │     no change │
│ QQuery 40 │   13.88 ms │                                  13.73 ms │     no change │
│ QQuery 41 │   12.65 ms │                                  13.12 ms │     no change │
│ QQuery 42 │   12.05 ms │                                  12.80 ms │  1.06x slower │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 40458.82ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 39736.75ms │
│ Average Time (HEAD)                                      │   940.90ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   924.11ms │
│ Queries Faster                                           │          6 │
│ Queries Slower                                           │          1 │
│ Queries with No Change                                   │         36 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-t-kid_single-dictionary-group-values
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                   HEAD ┃ rich-t-kid_single-dictionary-group-values ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │            1.15 / 1.78 ±1.15 / 4.08 ms │               1.11 / 1.76 ±1.18 / 4.12 ms │     no change │
│ QQuery 1  │         11.72 / 11.83 ±0.10 / 12.01 ms │            11.54 / 11.72 ±0.12 / 11.85 ms │     no change │
│ QQuery 2  │         34.89 / 35.13 ±0.20 / 35.44 ms │            34.54 / 35.69 ±0.85 / 36.91 ms │     no change │
│ QQuery 3  │         30.65 / 31.81 ±0.64 / 32.45 ms │            30.28 / 30.85 ±0.38 / 31.38 ms │     no change │
│ QQuery 4  │      236.74 / 245.68 ±6.65 / 254.02 ms │         246.74 / 257.01 ±8.84 / 270.98 ms │     no change │
│ QQuery 5  │     833.84 / 860.50 ±20.59 / 888.00 ms │         738.85 / 755.71 ±8.52 / 761.38 ms │ +1.14x faster │
│ QQuery 6  │            1.26 / 1.43 ±0.24 / 1.90 ms │               1.15 / 1.31 ±0.23 / 1.76 ms │ +1.09x faster │
│ QQuery 7  │         13.09 / 13.24 ±0.15 / 13.51 ms │            12.84 / 15.33 ±4.36 / 24.04 ms │  1.16x slower │
│ QQuery 8  │      336.96 / 342.71 ±5.80 / 353.57 ms │         342.02 / 350.64 ±7.30 / 363.63 ms │     no change │
│ QQuery 9  │     481.34 / 496.71 ±11.20 / 515.89 ms │         489.34 / 495.74 ±5.22 / 504.30 ms │     no change │
│ QQuery 10 │      203.18 / 204.06 ±1.10 / 206.22 ms │         200.24 / 203.10 ±1.68 / 205.42 ms │     no change │
│ QQuery 11 │      214.23 / 215.57 ±1.27 / 217.98 ms │         212.57 / 215.30 ±2.06 / 217.99 ms │     no change │
│ QQuery 12 │     835.73 / 850.69 ±10.37 / 864.60 ms │        782.00 / 793.31 ±16.69 / 825.53 ms │ +1.07x faster │
│ QQuery 13 │  1108.82 / 1149.22 ±23.65 / 1180.68 ms │     1100.79 / 1117.23 ±11.36 / 1130.84 ms │     no change │
│ QQuery 14 │     867.00 / 890.80 ±18.20 / 913.95 ms │        876.54 / 904.48 ±20.96 / 926.47 ms │     no change │
│ QQuery 15 │     296.42 / 310.75 ±11.90 / 331.26 ms │        305.55 / 319.89 ±10.79 / 337.97 ms │     no change │
│ QQuery 16 │  1345.32 / 1375.89 ±22.80 / 1413.99 ms │     1378.29 / 1404.76 ±25.70 / 1451.12 ms │     no change │
│ QQuery 17 │  1305.27 / 1328.74 ±18.33 / 1360.70 ms │     1341.38 / 1372.87 ±28.61 / 1421.85 ms │     no change │
│ QQuery 18 │  2101.48 / 2117.99 ±10.55 / 2134.47 ms │     2107.64 / 2149.02 ±34.67 / 2202.51 ms │     no change │
│ QQuery 19 │         28.48 / 29.16 ±0.63 / 30.35 ms │            28.80 / 29.24 ±0.41 / 29.96 ms │     no change │
│ QQuery 20 │  1063.35 / 1081.78 ±12.44 / 1101.46 ms │     1056.67 / 1080.14 ±19.78 / 1106.45 ms │     no change │
│ QQuery 21 │  1366.70 / 1388.58 ±19.12 / 1423.44 ms │      1385.05 / 1396.30 ±5.94 / 1402.78 ms │     no change │
│ QQuery 22 │  3270.97 / 3321.03 ±45.16 / 3403.61 ms │     3297.05 / 3352.86 ±43.62 / 3430.97 ms │     no change │
│ QQuery 23 │ 9236.77 / 9478.60 ±141.21 / 9673.84 ms │    9376.38 / 9635.28 ±197.95 / 9869.96 ms │     no change │
│ QQuery 24 │       97.23 / 102.33 ±4.37 / 109.30 ms │          96.21 / 100.02 ±2.88 / 103.79 ms │     no change │
│ QQuery 25 │     356.23 / 365.32 ±11.54 / 387.74 ms │         358.03 / 359.84 ±1.58 / 362.68 ms │     no change │
│ QQuery 26 │        98.45 / 99.65 ±1.30 / 101.98 ms │           97.28 / 98.90 ±1.24 / 100.00 ms │     no change │
│ QQuery 27 │   1157.18 / 1167.61 ±7.41 / 1177.04 ms │      1158.31 / 1171.10 ±9.76 / 1188.04 ms │     no change │
│ QQuery 28 │  2572.88 / 2614.16 ±26.30 / 2645.09 ms │     2519.63 / 2542.08 ±11.96 / 2555.00 ms │     no change │
│ QQuery 29 │         38.72 / 42.70 ±7.13 / 56.95 ms │            37.65 / 42.24 ±8.98 / 60.20 ms │     no change │
│ QQuery 30 │     557.93 / 571.52 ±11.21 / 585.13 ms │        558.05 / 575.86 ±27.41 / 630.35 ms │     no change │
│ QQuery 31 │     556.94 / 569.55 ±11.40 / 589.49 ms │        549.99 / 567.80 ±11.04 / 579.43 ms │     no change │
│ QQuery 32 │  1090.10 / 1145.12 ±44.56 / 1221.47 ms │     1088.35 / 1132.66 ±41.71 / 1211.07 ms │     no change │
│ QQuery 33 │  3887.48 / 3928.21 ±30.02 / 3976.43 ms │     3356.35 / 3543.02 ±94.86 / 3616.98 ms │ +1.11x faster │
│ QQuery 34 │  3854.98 / 3917.30 ±33.81 / 3951.90 ms │     3587.70 / 3648.90 ±76.89 / 3792.48 ms │ +1.07x faster │
│ QQuery 35 │     303.45 / 335.31 ±34.50 / 402.43 ms │        318.28 / 333.79 ±11.60 / 350.93 ms │     no change │
│ QQuery 36 │     135.90 / 162.28 ±21.65 / 200.63 ms │        123.67 / 150.62 ±17.89 / 175.87 ms │ +1.08x faster │
│ QQuery 37 │       97.46 / 100.69 ±3.55 / 107.11 ms │           95.05 / 99.71 ±3.77 / 105.23 ms │     no change │
│ QQuery 38 │         83.81 / 86.94 ±4.08 / 94.51 ms │            80.40 / 88.74 ±5.19 / 94.30 ms │     no change │
│ QQuery 39 │     306.14 / 333.65 ±14.69 / 345.77 ms │        314.79 / 334.46 ±14.76 / 349.95 ms │     no change │
│ QQuery 40 │         13.88 / 14.41 ±0.46 / 15.05 ms │            13.73 / 14.88 ±0.65 / 15.70 ms │     no change │
│ QQuery 41 │         12.65 / 13.17 ±0.75 / 14.67 ms │            13.12 / 16.98 ±6.63 / 30.18 ms │  1.29x slower │
│ QQuery 42 │         12.05 / 12.34 ±0.22 / 12.64 ms │            12.80 / 13.09 ±0.25 / 13.52 ms │  1.06x slower │
└───────────┴────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 41365.95ms │
│ Total Time (rich-t-kid_single-dictionary-group-values)   │ 40764.23ms │
│ Average Time (HEAD)                                      │   962.00ms │
│ Average Time (rich-t-kid_single-dictionary-group-values) │   948.01ms │
│ Queries Faster                                           │          6 │
│ Queries Slower                                           │          3 │
│ Queries with No Change                                   │         34 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 210.0s
Peak memory 13.0 GiB
Avg memory 5.6 GiB
CPU user 2014.2s
CPU sys 134.8s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 205.0s
Peak memory 12.7 GiB
Avg memory 5.4 GiB
CPU user 1951.0s
CPU sys 138.3s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing rich-t-kid/single-dictionary-group-values (3f7f357) to c8265b6 (merge-base) diff

Run configuration
run benchmark dictionary_group_values
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

group                                                                             HEAD                                    rich-t-kid_single-dictionary-group-values
-----                                                                             ----                                    -----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.91    481.5±4.82µs 129.8 MElem/sec    1.00    252.4±1.37µs 247.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.97    416.4±3.11µs 150.1 MElem/sec    1.00    211.8±1.22µs 295.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.98    441.4±3.76µs 141.6 MElem/sec    1.00    222.8±1.88µs 280.6 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.90      6.0±0.01ms 10.4 MElem/sec     1.00      3.2±0.01ms 19.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         2.03    433.7±3.58µs 144.1 MElem/sec    1.00    213.8±1.39µs 292.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.66    112.1±0.64µs 69.7 MElem/sec     1.00     67.4±0.46µs 115.9 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.97     56.2±0.85µs 139.0 MElem/sec    1.00     28.5±0.52µs 274.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.83     72.4±0.77µs 107.9 MElem/sec    1.00     39.6±0.50µs 197.5 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          2.00     61.4±0.77µs 127.3 MElem/sec    1.00     30.6±0.57µs 255.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.49    521.7±1.97µs 15.0 MElem/sec     1.00    349.3±1.79µs 22.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.64      2.2±0.00ms 112.9 MElem/sec    1.00   1347.9±2.14µs 185.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.68      2.0±0.00ms 122.1 MElem/sec    1.00   1220.2±2.04µs 204.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.69      2.1±0.00ms 117.7 MElem/sec    1.00   1255.3±2.40µs 199.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.64     18.9±0.19ms 13.2 MElem/sec     1.00     11.6±0.28ms 21.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.73      2.1±0.00ms 118.0 MElem/sec    1.00   1227.6±4.69µs 203.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.41    378.6±1.90µs 82.5 MElem/sec     1.00    269.0±1.21µs 116.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.72    249.8±1.21µs 125.1 MElem/sec    1.00    145.4±0.49µs 215.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.60    286.5±1.50µs 109.1 MElem/sec    1.00    178.6±0.78µs 174.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.82  277.0±150.57µs 112.8 MElem/sec    1.00    152.1±0.61µs 205.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.26   1289.1±4.75µs 24.2 MElem/sec     1.00   1022.8±9.99µs 30.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       3.81   1339.0±4.40µs 46.7 MElem/sec     1.00    351.1±1.28µs 178.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         4.06   1210.6±3.01µs 51.6 MElem/sec     1.00    298.5±1.37µs 209.4 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        4.00   1245.0±3.29µs 50.2 MElem/sec     1.00    311.1±0.76µs 200.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      2.03      7.5±0.02ms  8.4 MElem/sec     1.00      3.7±0.03ms 17.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         4.10   1232.1±3.69µs 50.7 MElem/sec     1.00    300.2±0.73µs 208.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    2.61     18.9±0.17ms 13.2 MElem/sec     1.00      7.3±0.02ms 34.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     2.92      7.8±0.03ms 32.1 MElem/sec     1.00      2.7±0.01ms 93.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    3.01     63.8±2.16ms  3.9 MElem/sec     1.00     21.2±0.34ms 11.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     2.57     14.0±0.08ms 17.9 MElem/sec     1.00      5.4±0.01ms 45.9 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      2.64      2.8±0.01ms 90.3 MElem/sec     1.00   1049.3±4.19µs 238.3 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.3±0.01ms 106.7 MElem/sec    1.02      2.4±0.00ms 105.0 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.0±0.00ms 124.8 MElem/sec    1.08      2.2±0.01ms 115.3 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.1±0.01ms 117.6 MElem/sec    1.05      2.2±0.00ms 111.8 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.89     25.9±0.08ms  9.7 MElem/sec     1.00     13.7±0.08ms 18.2 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.1±0.00ms 120.5 MElem/sec    1.06      2.2±0.00ms 113.9 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric Value
Wall time 420.1s
Peak memory 901.3 MiB
Avg memory 132.1 MiB
CPU user 410.0s
CPU sys 13.9s
Peak spill 0 B

dictionary_group_values — branch

Metric Value
Wall time 520.1s
Peak memory 548.7 MiB
Avg memory 381.2 MiB
CPU user 449.6s
CPU sys 0.5s
Peak spill 0 B

File an issue against this benchmark runner

@kumarUjjawal
kumarUjjawal self-requested a review November 13, 2026 04:53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

physical-plan Changes to the physical-plan crate

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants