DRAFT Bench 25413 by kosiew · Pull Request #26080 · apache/datafusion

kosiew · 2026-10-06T10:07:28Z

Run benchmarks

kosiew · 2026-10-06T10:08:16Z

run benchmark nlj

adriangbot · 2026-10-06T10:18:06Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highcpu-48 (12 vCPU / 65 GiB) | Linux bench-c6014003302-3090-lntj9 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  48
On-line CPU(s) list:                     0-47
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     48
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:                               3 MiB (48 instances)
L1i cache:                               3 MiB (48 instances)
L2 cache:                                96 MiB (48 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-47
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 bench-25413 (e7545a0) to 827ca29 (merge-base) diff

Run configuration

run benchmark nlj

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-10-06T10:29:07Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highcpu-48 (12 vCPU / 65 GiB)

Comparing bench-25413 (e7545a0) to 827ca29 (merge-base) diff

Run configuration

run benchmark nlj

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  48
On-line CPU(s) list:                     0-47
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     48
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:                               3 MiB (48 instances)
L1i cache:                               3 MiB (48 instances)
L2 cache:                                96 MiB (48 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-47
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 bench-25413
--------------------
Benchmark nlj.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ bench-25413 ┃          Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩
│ QQuery 1  │  254.46 ms │   254.80 ms │       no change │
│ QQuery 2  │  314.32 ms │   314.10 ms │       no change │
│ QQuery 3  │  389.43 ms │   395.75 ms │       no change │
│ QQuery 4  │ 1026.63 ms │  1024.77 ms │       no change │
│ QQuery 5  │  640.13 ms │   642.31 ms │       no change │
│ QQuery 6  │ 5287.59 ms │  5293.40 ms │       no change │
│ QQuery 7  │  649.72 ms │   645.36 ms │       no change │
│ QQuery 8  │ 5295.99 ms │  5306.96 ms │       no change │
│ QQuery 9  │  777.60 ms │   779.32 ms │       no change │
│ QQuery 10 │ 1221.46 ms │  1234.21 ms │       no change │
│ QQuery 11 │  324.36 ms │   327.97 ms │       no change │
│ QQuery 12 │  327.99 ms │   332.83 ms │       no change │
│ QQuery 13 │  111.08 ms │   110.61 ms │       no change │
│ QQuery 14 │  110.75 ms │   111.00 ms │       no change │
│ QQuery 15 │  110.09 ms │     0.58 ms │ +188.66x faster │
│ QQuery 16 │  110.38 ms │     0.59 ms │ +185.80x faster │
│ QQuery 17 │  113.08 ms │   112.37 ms │       no change │
└───────────┴────────────┴─────────────┴─────────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary          ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)          │ 17065.03ms │
│ Total Time (bench-25413)   │ 16886.94ms │
│ Average Time (HEAD)        │  1003.83ms │
│ Average Time (bench-25413) │   993.35ms │
│ Queries Faster             │          2 │
│ Queries Slower             │          0 │
│ Queries with No Change     │         15 │
│ Queries with Failure       │          0 │
└────────────────────────────┴────────────┘

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

Comparing HEAD and bench-25413
--------------------
Benchmark nlj.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                           bench-25413 ┃          Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩
│ QQuery 1  │     254.46 / 254.73 ±0.22 / 255.02 ms │     254.80 / 255.34 ±0.72 / 256.77 ms │       no change │
│ QQuery 2  │     314.32 / 317.44 ±2.47 / 319.56 ms │     314.10 / 314.33 ±0.33 / 314.96 ms │       no change │
│ QQuery 3  │     389.43 / 391.18 ±1.83 / 394.40 ms │     395.75 / 396.95 ±0.66 / 397.67 ms │       no change │
│ QQuery 4  │  1026.63 / 1028.35 ±1.05 / 1029.64 ms │  1024.77 / 1027.55 ±2.40 / 1031.30 ms │       no change │
│ QQuery 5  │     640.13 / 642.40 ±2.04 / 646.23 ms │     642.31 / 643.36 ±1.08 / 645.28 ms │       no change │
│ QQuery 6  │ 5287.59 / 5307.95 ±19.96 / 5340.86 ms │ 5293.40 / 5309.35 ±12.55 / 5331.44 ms │       no change │
│ QQuery 7  │     649.72 / 650.98 ±0.64 / 651.56 ms │     645.36 / 646.00 ±0.43 / 646.69 ms │       no change │
│ QQuery 8  │  5295.99 / 5302.88 ±4.50 / 5308.02 ms │ 5306.96 / 5326.48 ±16.85 / 5350.14 ms │       no change │
│ QQuery 9  │     777.60 / 781.11 ±2.07 / 783.00 ms │     779.32 / 782.30 ±2.52 / 785.49 ms │       no change │
│ QQuery 10 │  1221.46 / 1228.36 ±5.99 / 1239.20 ms │  1234.21 / 1243.16 ±6.13 / 1250.10 ms │       no change │
│ QQuery 11 │     324.36 / 325.69 ±0.70 / 326.36 ms │     327.97 / 331.27 ±2.30 / 334.94 ms │       no change │
│ QQuery 12 │     327.99 / 331.60 ±1.95 / 333.91 ms │     332.83 / 336.20 ±2.96 / 341.68 ms │       no change │
│ QQuery 13 │     111.08 / 111.86 ±0.67 / 112.86 ms │     110.61 / 111.82 ±0.75 / 112.70 ms │       no change │
│ QQuery 14 │     110.75 / 112.09 ±1.11 / 113.49 ms │     111.00 / 112.66 ±1.00 / 113.76 ms │       no change │
│ QQuery 15 │     110.09 / 110.52 ±0.38 / 111.15 ms │           0.58 / 0.66 ±0.05 / 0.73 ms │ +167.81x faster │
│ QQuery 16 │     110.38 / 110.86 ±0.28 / 111.22 ms │           0.59 / 0.62 ±0.01 / 0.64 ms │ +180.19x faster │
│ QQuery 17 │     113.08 / 114.08 ±0.65 / 114.95 ms │     112.37 / 113.88 ±0.99 / 115.05 ms │       no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴─────────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary          ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)          │ 17122.07ms │
│ Total Time (bench-25413)   │ 16951.93ms │
│ Average Time (HEAD)        │  1007.18ms │
│ Average Time (bench-25413) │   997.17ms │
│ Queries Faster             │          2 │
│ Queries Slower             │          0 │
│ Queries with No Change     │         15 │
│ Queries with Failure       │          0 │
└────────────────────────────┴────────────┘

Resource Usage

nlj — base (merge-base)

Metric	Value
Wall time	90.0s
Peak memory	70.8 MiB
Avg memory	60.6 MiB
CPU user	209.6s
CPU sys	3.3s
Peak spill	0 B

nlj — branch

Metric	Value
Wall time	85.0s
Peak memory	67.9 MiB
Avg memory	62.8 MiB
CPU user	205.6s
CPU sys	3.3s
Peak spill	0 B

File an issue against this benchmark runner

codecov-commenter · 2026-10-06T10:31:58Z

Codecov Report

❌ Patch coverage is 87.86408% with 25 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.72%. Comparing base (827ca29) to head (e7545a0).
⚠️ Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
...fusion/physical-plan/src/joins/nested_loop_join.rs	87.86%	10 Missing and 15 partials ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##             main   #26080    +/-   ##
========================================
  Coverage   82.71%   82.72%            
========================================
  Files        1147     1147            
  Lines      447818   448198   +380     
  Branches   447818   448198   +380     
========================================
+ Hits       370412   370759   +347     
- Misses      54919    54935    +16     
- Partials    22487    22504    +17

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

lyne7-sc and others added 3 commits September 17, 2026 21:25

perf: prune matched candidates in right semi/anti nested loop joins

b9fb069

Merge upstream/main into perf/nlj-right-existence-pruning

0c09e33

Merge branch 'main' into perf/nlj-right-existence-pruning

9d19d30

kosiew changed the title ~~Bench 25413~~ DRAFT Bench 25413 Oct 6, 2026

github-actions Bot added sqllogictest SQL Logic Tests (.slt) physical-plan Changes to the physical-plan crate labels Oct 6, 2026

kosiew mentioned this pull request Oct 6, 2026

perf: Prune matched right rows in nested loop semi and anti joins #25413

Open

kosiew force-pushed the bench-25413 branch from e7545a0 to 9d19d30 Compare October 6, 2026 11:25

Navigation Menu

Repository navigation

DRAFT Bench 25413 - #26080

DRAFT Bench 25413 #26080
kosiew wants to merge 3 commits into
apache:mainfrom
kosiew:bench-25413

kosiew commented Oct 6, 2026 •

edited

Loading

Uh oh!

kosiew commented Oct 6, 2026

Uh oh!

adriangbot commented Oct 6, 2026

Uh oh!

adriangbot commented Oct 6, 2026

Uh oh!

codecov-commenter commented Oct 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

kosiew commented Oct 6, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kosiew commented Oct 6, 2026

Uh oh!

adriangbot commented Oct 6, 2026

Uh oh!

adriangbot commented Oct 6, 2026

Uh oh!

codecov-commenter commented Oct 6, 2026

Codecov Report

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

kosiew commented Oct 6, 2026 •

edited

Loading