You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "Rich-T-kid (via GitHub)" <gi...@apache.org> on 2026/10/13 18:22:21 UTC
[PR] Fix dictionary LIKE coercion [datafusion]
Rich-T-kid opened a new pull request, #26279:
URL: https://github.com/apache/datafusion/pull/26279
## Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and enhancements and this helps us generate change logs for our releases. You can link an issue to this PR using the GitHub syntax. For example `Closes #123` indicates that this PR will close issue #123.
-->
- Closes #.
## Rationale for this change
<!--
Why are you proposing this change? If this is already explained clearly in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand your changes and offer better suggestions for fixes.
Please explain the problem you are trying to solve in terms of the user-visible
behavior, rather than the implementation.
For example, "The code in `foo.rs` doesn't handle nulls" is a symptom of the
implementation. "COUNT(DISTINCT) returns wrong results when the column contains
nulls" is the user-visible problem.
-->
## What changes are included in this PR?
<!--
There is no need to duplicate the description in the issue here, but it is sometimes worth providing a summary of the individual changes in this PR.
-->
## What is the testing strategy for this PR?
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
Briefly describe how this PR is tested, and point to the specific tests you added. For example: 'This new feature is covered by the `sqllogictest` cases added in `foo.slt`'.
If this PR does not add tests, explain why. For example, if the change is already covered by existing tests, please mention it.
You should also check the `codecov` bot reply on this PR to confirm the changed code is exercised.
-->
## Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be updated before approving the PR.
If there are any breaking changes to public APIs, please add the `api change` label.
-->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "Rich-T-kid (via GitHub)" <gi...@apache.org>.
Rich-T-kid closed pull request #26279: Fix dictionary LIKE coercion
URL: https://github.com/apache/datafusion/pull/26279
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152527291
π€ Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/26279#issuecomment-6152371026)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
Comparing rich-T-kid/dict-like-preserve (6ee6be26b6cc897eda6961aece9135aa2aab50a0) to 2c568b4 (merge-base) [diff](https://github.com/apache/datafusion/compare/2c568b478b3aff18de52609c7237f1c3f494d5b1..6ee6be26b6cc897eda6961aece9135aa2aab50a0)
<details><summary>Run configuration</summary>
```yaml
run benchmark tpch
baseline:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
changed:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
```
</details>
<details><summary>CPU Details (lscpu)</summary>
```
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>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and rich-T-kid_dict-like-preserve
--------------------
Benchmark tpch_sf1.json
--------------------
βββββββββββββ³βββββββββββ³ββββββββββββββββββββββββββββββββ³βββββββββββββββ
β Query β HEAD β rich-T-kid_dict-like-preserve β Change β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 1 β 42.13 ms β 43.47 ms β no change β
β QQuery 2 β 20.54 ms β 20.89 ms β no change β
β QQuery 3 β 31.87 ms β 31.83 ms β no change β
β QQuery 4 β 19.22 ms β 18.85 ms β no change β
β QQuery 5 β 40.04 ms β 39.89 ms β no change β
β QQuery 6 β 14.61 ms β 14.81 ms β no change β
β QQuery 7 β 45.77 ms β 46.13 ms β no change β
β QQuery 8 β 43.61 ms β 44.46 ms β no change β
β QQuery 9 β 54.89 ms β 55.27 ms β no change β
β QQuery 10 β 62.99 ms β 63.85 ms β no change β
β QQuery 11 β 14.64 ms β 14.88 ms β no change β
β QQuery 12 β 21.92 ms β 22.94 ms β no change β
β QQuery 13 β 50.04 ms β 54.30 ms β 1.09x slower β
β QQuery 14 β 26.16 ms β 27.31 ms β no change β
β QQuery 15 β 33.67 ms β 34.90 ms β no change β
β QQuery 16 β 15.16 ms β 15.25 ms β no change β
β QQuery 17 β 86.32 ms β 89.45 ms β no change β
β QQuery 18 β 73.17 ms β 74.62 ms β no change β
β QQuery 19 β 30.78 ms β 32.14 ms β no change β
β QQuery 20 β 35.32 ms β 38.47 ms β 1.09x slower β
β QQuery 21 β 59.81 ms β 61.47 ms β no change β
β QQuery 22 β 17.89 ms β 17.34 ms β no change β
βββββββββββββ΄βββββββββββ΄ββββββββββββββββββββββββββββββββ΄βββββββββββββββ
ββββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββ
β Benchmark Summary β β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 840.56ms β
β Total Time (rich-T-kid_dict-like-preserve) β 862.49ms β
β Average Time (HEAD) β 38.21ms β
β Average Time (rich-T-kid_dict-like-preserve) β 39.20ms β
β Queries Faster β 0 β
β Queries Slower β 2 β
β Queries with No Change β 20 β
β Queries with Failure β 0 β
ββββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββ
Distribution per query (min / mean Β±stddev / max):
Comparing HEAD and rich-T-kid_dict-like-preserve
--------------------
Benchmark tpch_sf1.json
--------------------
βββββββββββββ³βββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββ³βββββββββββββββ
β Query β HEAD β rich-T-kid_dict-like-preserve β Change β
β‘βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 1 β 42.13 / 44.31 Β±2.53 / 48.99 ms β 43.47 / 44.97 Β±2.47 / 49.90 ms β no change β
β QQuery 2 β 20.54 / 21.03 Β±0.40 / 21.72 ms β 20.89 / 21.30 Β±0.38 / 21.90 ms β no change β
β QQuery 3 β 31.87 / 32.35 Β±0.30 / 32.74 ms β 31.83 / 32.68 Β±1.27 / 35.20 ms β no change β
β QQuery 4 β 19.22 / 19.77 Β±0.69 / 21.13 ms β 18.85 / 19.08 Β±0.14 / 19.28 ms β no change β
β QQuery 5 β 40.04 / 40.36 Β±0.22 / 40.60 ms β 39.89 / 41.11 Β±1.35 / 43.15 ms β no change β
β QQuery 6 β 14.61 / 14.80 Β±0.13 / 14.95 ms β 14.81 / 15.12 Β±0.43 / 15.97 ms β no change β
β QQuery 7 β 45.77 / 48.32 Β±1.52 / 49.61 ms β 46.13 / 48.39 Β±1.67 / 50.24 ms β no change β
β QQuery 8 β 43.61 / 44.38 Β±0.81 / 45.93 ms β 44.46 / 44.65 Β±0.14 / 44.85 ms β no change β
β QQuery 9 β 54.89 / 56.03 Β±1.25 / 58.47 ms β 55.27 / 56.32 Β±1.24 / 58.69 ms β no change β
β QQuery 10 β 62.99 / 63.90 Β±0.97 / 65.60 ms β 63.85 / 65.05 Β±0.75 / 65.75 ms β no change β
β QQuery 11 β 14.64 / 15.26 Β±0.93 / 17.11 ms β 14.88 / 16.18 Β±1.84 / 19.80 ms β 1.06x slower β
β QQuery 12 β 21.92 / 22.29 Β±0.41 / 23.09 ms β 22.94 / 23.25 Β±0.40 / 23.98 ms β no change β
β QQuery 13 β 50.04 / 50.77 Β±0.49 / 51.33 ms β 54.30 / 55.59 Β±0.93 / 56.67 ms β 1.09x slower β
β QQuery 14 β 26.16 / 26.40 Β±0.15 / 26.59 ms β 27.31 / 27.52 Β±0.27 / 28.04 ms β no change β
β QQuery 15 β 33.67 / 34.50 Β±0.52 / 35.18 ms β 34.90 / 35.65 Β±0.84 / 37.09 ms β no change β
β QQuery 16 β 15.16 / 15.38 Β±0.13 / 15.57 ms β 15.25 / 15.61 Β±0.24 / 15.88 ms β no change β
β QQuery 17 β 86.32 / 87.31 Β±0.80 / 88.70 ms β 89.45 / 91.88 Β±3.48 / 98.67 ms β 1.05x slower β
β QQuery 18 β 73.17 / 77.28 Β±2.95 / 80.86 ms β 74.62 / 75.95 Β±1.12 / 77.93 ms β no change β
β QQuery 19 β 30.78 / 31.33 Β±0.51 / 32.22 ms β 32.14 / 33.21 Β±0.88 / 34.79 ms β 1.06x slower β
β QQuery 20 β 35.32 / 35.97 Β±0.54 / 36.59 ms β 38.47 / 38.81 Β±0.41 / 39.36 ms β 1.08x slower β
β QQuery 21 β 59.81 / 63.75 Β±2.70 / 67.27 ms β 61.47 / 61.98 Β±0.41 / 62.60 ms β no change β
β QQuery 22 β 17.89 / 17.99 Β±0.11 / 18.19 ms β 17.34 / 17.45 Β±0.06 / 17.50 ms β no change β
βββββββββββββ΄βββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββ΄βββββββββββββββ
ββββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββ
β Benchmark Summary β β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 863.47ms β
β Total Time (rich-T-kid_dict-like-preserve) β 881.74ms β
β Average Time (HEAD) β 39.25ms β
β Average Time (rich-T-kid_dict-like-preserve) β 40.08ms β
β Queries Faster β 0 β
β Queries Slower β 5 β
β Queries with No Change β 17 β
β Queries with Failure β 0 β
ββββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββ
```
</p>
</details>
<details><summary>Resource Usage</summary>
**tpch β base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 5.0s |
| Peak memory | 1.1 GiB |
| Avg memory | 634.8 MiB |
| CPU user | 21.7s |
| CPU sys | 1.7s |
| Peak spill | 0 B |
**tpch β branch**
| Metric | Value |
|--------|-------|
| Wall time | 5.0s |
| Peak memory | 1.1 GiB |
| Avg memory | 625.8 MiB |
| CPU user | 21.5s |
| CPU sys | 1.9s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "Rich-T-kid (via GitHub)" <gi...@apache.org>.
Rich-T-kid commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152344246
https://github.com/apache/datafusion/pull/24227#issuecomment-5675290885
^ will be useful in the future
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152398014
π€ Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/26279#issuecomment-6152377986)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c6152377986-3368-9hjv6 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
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 rich-T-kid/dict-like-preserve (6ee6be26b6cc897eda6961aece9135aa2aab50a0) to 2c568b4 (merge-base) [diff](https://github.com/apache/datafusion/compare/2c568b478b3aff18de52609c7237f1c3f494d5b1..6ee6be26b6cc897eda6961aece9135aa2aab50a0)
<details><summary>Run configuration</summary>
```yaml
run benchmark clickbench_partitioned
baseline:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
changed:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
```
</details>
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "Rich-T-kid (via GitHub)" <gi...@apache.org>.
Rich-T-kid commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152377986
run benchmark clickbench_partitioned
baseline:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
changed:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "codecov-commenter (via GitHub)" <gi...@apache.org>.
codecov-commenter commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6157405060
## [Codecov](https://app.codecov.io/gh/apache/datafusion/pull/26279?dropdown=coverage&src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) Report
:x: Patch coverage is `80.95238%` with `4 lines` in your changes missing coverage. Please review.
:white_check_mark: Project coverage is 82.83%. Comparing base ([`2c568b4`](https://app.codecov.io/gh/apache/datafusion/commit/2c568b478b3aff18de52609c7237f1c3f494d5b1?dropdown=coverage&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache)) to head ([`6ee6be2`](https://app.codecov.io/gh/apache/datafusion/commit/6ee6be26b6cc897eda6961aece9135aa2aab50a0?dropdown=coverage&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache)).
| [Files with missing lines](https://app.codecov.io/gh/apache/datafusion/pull/26279?dropdown=coverage&src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) | Patch % | Lines |
|---|---|---|
| [datafusion/optimizer/src/analyzer/type\_coercion.rs](https://app.codecov.io/gh/apache/datafusion/pull/26279?src=pr&el=tree&filepath=datafusion%2Foptimizer%2Fsrc%2Fanalyzer%2Ftype_coercion.rs&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#diff-ZGF0YWZ1c2lvbi9vcHRpbWl6ZXIvc3JjL2FuYWx5emVyL3R5cGVfY29lcmNpb24ucnM=) | 80.95% | [1 Missing and 3 partials :warning: ](https://app.codecov.io/gh/apache/datafusion/pull/26279?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) |
<details><summary>Additional details and impacted files</summary>
```diff
@@ Coverage Diff @@
## main #26279 +/- ##
==========================================
- Coverage 82.83% 82.83% -0.01%
==========================================
Files 1149 1149
Lines 452798 452814 +16
Branches 452798 452814 +16
==========================================
+ Hits 375082 375089 +7
- Misses 54893 54896 +3
- Partials 22823 22829 +6
```
</details>
[:umbrella: View full report in Codecov by Harness](https://app.codecov.io/gh/apache/datafusion/pull/26279?dropdown=coverage&src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache).
:loudspeaker: Have feedback on the report? [Share it here](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache).
<details><summary> :rocket: New features to boost your workflow: </summary>
- :snowflake: [Test Analytics](https://docs.codecov.com/docs/test-analytics): Detect flaky tests, report on failures, and find test suite problems.
- :package: [JS Bundle Analysis](https://docs.codecov.com/docs/javascript-bundle-analysis): Save yourself from yourself by tracking and limiting bundle sizes in JS merges.
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152390679
π€ Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/26279#issuecomment-6152371026)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c6152371026-3367-shg5j 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
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 rich-T-kid/dict-like-preserve (6ee6be26b6cc897eda6961aece9135aa2aab50a0) to 2c568b4 (merge-base) [diff](https://github.com/apache/datafusion/compare/2c568b478b3aff18de52609c7237f1c3f494d5b1..6ee6be26b6cc897eda6961aece9135aa2aab50a0)
<details><summary>Run configuration</summary>
```yaml
run benchmark tpch
baseline:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
changed:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
```
</details>
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "Rich-T-kid (via GitHub)" <gi...@apache.org>.
Rich-T-kid commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152371026
run benchmark tpch
```rust
baseline:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
changed:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Fix dictionary LIKE coercion [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26279:
URL: https://github.com/apache/datafusion/pull/26279#issuecomment-6152807000
π€ Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/26279#issuecomment-6152377986)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
Comparing rich-T-kid/dict-like-preserve (6ee6be26b6cc897eda6961aece9135aa2aab50a0) to 2c568b4 (merge-base) [diff](https://github.com/apache/datafusion/compare/2c568b478b3aff18de52609c7237f1c3f494d5b1..6ee6be26b6cc897eda6961aece9135aa2aab50a0)
<details><summary>Run configuration</summary>
```yaml
run benchmark clickbench_partitioned
baseline:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
changed:
env:
DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
```
</details>
<details><summary>CPU Details (lscpu)</summary>
```
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>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and rich-T-kid_dict-like-preserve
--------------------
Benchmark clickbench_partitioned.json
--------------------
βββββββββββββ³ββββββββββββββ³ββββββββββββββββββββββββββββββββ³βββββββββββββββ
β Query β HEAD β rich-T-kid_dict-like-preserve β Change β
β‘βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 0 β 1.27 ms β 1.25 ms β no change β
β QQuery 1 β 11.75 ms β 11.67 ms β no change β
β QQuery 2 β 36.00 ms β 36.37 ms β no change β
β QQuery 3 β 32.47 ms β 32.31 ms β no change β
β QQuery 4 β 270.25 ms β 270.39 ms β no change β
β QQuery 5 β 1608.96 ms β 1600.33 ms β no change β
β QQuery 6 β 1.39 ms β 1.46 ms β no change β
β QQuery 7 β 13.32 ms β 13.56 ms β no change β
β QQuery 8 β 382.87 ms β 376.30 ms β no change β
β QQuery 9 β 501.05 ms β 550.48 ms β 1.10x slower β
β QQuery 10 β 203.77 ms β 202.91 ms β no change β
β QQuery 11 β 213.01 ms β 213.88 ms β no change β
β QQuery 12 β 1663.83 ms β 1658.37 ms β no change β
β QQuery 13 β 2319.95 ms β 2254.81 ms β no change β
β QQuery 14 β 1800.29 ms β 1769.27 ms β no change β
β QQuery 15 β 325.11 ms β 327.37 ms β no change β
β QQuery 16 β 3904.79 ms β 3925.52 ms β no change β
β QQuery 17 β 4136.30 ms β 4105.94 ms β no change β
β QQuery 18 β 7645.62 ms β 7464.94 ms β no change β
β QQuery 19 β 29.15 ms β 29.47 ms β no change β
β QQuery 20 β 1119.11 ms β 1100.10 ms β no change β
β QQuery 21 β 1397.49 ms β 1427.18 ms β no change β
β QQuery 22 β 3322.79 ms β 3336.20 ms β no change β
β QQuery 23 β 9532.23 ms β 9616.71 ms β no change β
β QQuery 24 β 100.82 ms β 99.84 ms β no change β
β QQuery 25 β 359.50 ms β 358.58 ms β no change β
β QQuery 26 β 100.19 ms β 100.05 ms β no change β
β QQuery 27 β 1198.84 ms β 1193.35 ms β no change β
β QQuery 28 β 3622.83 ms β 3670.71 ms β no change β
β QQuery 29 β 40.70 ms β 40.52 ms β no change β
β QQuery 30 β 574.94 ms β 574.07 ms β no change β
β QQuery 31 β 572.41 ms β 556.12 ms β no change β
β QQuery 32 β 1128.50 ms β 1100.72 ms β no change β
β QQuery 33 β 14819.92 ms β 15108.04 ms β no change β
β QQuery 34 β 15465.47 ms β 15862.02 ms β no change β
β QQuery 35 β 358.01 ms β 363.71 ms β no change β
β QQuery 36 β 192.32 ms β 200.11 ms β no change β
β QQuery 37 β 98.94 ms β 100.27 ms β no change β
β QQuery 38 β 82.47 ms β 81.10 ms β no change β
β QQuery 39 β 431.00 ms β 453.91 ms β 1.05x slower β
β QQuery 40 β 15.09 ms β 16.71 ms β 1.11x slower β
β QQuery 41 β 14.43 ms β 14.79 ms β no change β
β QQuery 42 β 14.12 ms β 14.53 ms β no change β
βββββββββββββ΄ββββββββββββββ΄ββββββββββββββββββββββββββββββββ΄βββββββββββββββ
ββββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββ
β Benchmark Summary β β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 79663.28ms β
β Total Time (rich-T-kid_dict-like-preserve) β 80235.89ms β
β Average Time (HEAD) β 1852.63ms β
β Average Time (rich-T-kid_dict-like-preserve) β 1865.95ms β
β Queries Faster β 0 β
β Queries Slower β 3 β
β Queries with No Change β 40 β
β Queries with Failure β 0 β
ββββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββ
Distribution per query (min / mean Β±stddev / max):
Comparing HEAD and rich-T-kid_dict-like-preserve
--------------------
Benchmark clickbench_partitioned.json
--------------------
βββββββββββββ³ββββββββββββββββββββββββββββββββββββββββββββ³ββββββββββββββββββββββββββββββββββββββββββββ³ββββββββββββββββ
β Query β HEAD β rich-T-kid_dict-like-preserve β Change β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 0 β 1.27 / 4.79 Β±6.76 / 18.30 ms β 1.25 / 4.72 Β±6.66 / 18.04 ms β no change β
β QQuery 1 β 11.75 / 12.07 Β±0.22 / 12.35 ms β 11.67 / 11.98 Β±0.25 / 12.34 ms β no change β
β QQuery 2 β 36.00 / 36.61 Β±0.50 / 37.29 ms β 36.37 / 36.89 Β±0.36 / 37.39 ms β no change β
β QQuery 3 β 32.47 / 33.54 Β±1.31 / 35.94 ms β 32.31 / 32.75 Β±0.54 / 33.81 ms β no change β
β QQuery 4 β 270.25 / 274.95 Β±3.53 / 279.62 ms β 270.39 / 273.16 Β±2.55 / 277.92 ms β no change β
β QQuery 5 β 1608.96 / 1720.17 Β±106.04 / 1898.96 ms β 1600.33 / 1699.27 Β±59.54 / 1757.91 ms β no change β
β QQuery 6 β 1.39 / 1.64 Β±0.26 / 1.97 ms β 1.46 / 1.65 Β±0.20 / 2.03 ms β no change β
β QQuery 7 β 13.32 / 13.58 Β±0.23 / 13.94 ms β 13.56 / 13.67 Β±0.08 / 13.76 ms β no change β
β QQuery 8 β 382.87 / 393.48 Β±14.52 / 421.90 ms β 376.30 / 388.71 Β±10.12 / 405.67 ms β no change β
β QQuery 9 β 501.05 / 541.42 Β±25.81 / 567.83 ms β 550.48 / 574.79 Β±16.13 / 596.72 ms β 1.06x slower β
β QQuery 10 β 203.77 / 203.85 Β±0.05 / 203.92 ms β 202.91 / 209.93 Β±11.93 / 233.73 ms β no change β
β QQuery 11 β 213.01 / 215.58 Β±2.11 / 217.97 ms β 213.88 / 215.70 Β±1.31 / 217.06 ms β no change β
β QQuery 12 β 1663.83 / 1725.16 Β±56.00 / 1809.11 ms β 1658.37 / 1690.82 Β±21.03 / 1720.23 ms β no change β
β QQuery 13 β 2319.95 / 2613.33 Β±339.80 / 3201.11 ms β 2254.81 / 2603.03 Β±258.40 / 2930.98 ms β no change β
β QQuery 14 β 1800.29 / 1851.02 Β±42.06 / 1913.13 ms β 1769.27 / 1836.71 Β±54.20 / 1923.45 ms β no change β
β QQuery 15 β 325.11 / 330.95 Β±4.24 / 337.98 ms β 327.37 / 347.80 Β±13.28 / 365.96 ms β 1.05x slower β
β QQuery 16 β 3904.79 / 4146.25 Β±143.69 / 4317.80 ms β 3925.52 / 4086.15 Β±145.97 / 4331.41 ms β no change β
β QQuery 17 β 4136.30 / 4236.65 Β±81.75 / 4357.10 ms β 4105.94 / 4208.32 Β±89.12 / 4354.59 ms β no change β
β QQuery 18 β 7645.62 / 7804.42 Β±194.76 / 8150.80 ms β 7464.94 / 7668.66 Β±158.39 / 7850.27 ms β no change β
β QQuery 19 β 29.15 / 29.59 Β±0.34 / 30.10 ms β 29.47 / 30.86 Β±2.27 / 35.36 ms β no change β
β QQuery 20 β 1119.11 / 1130.96 Β±8.06 / 1144.33 ms β 1100.10 / 1102.18 Β±1.59 / 1104.74 ms β no change β
β QQuery 21 β 1397.49 / 1433.57 Β±19.11 / 1452.66 ms β 1427.18 / 1452.14 Β±16.64 / 1472.64 ms β no change β
β QQuery 22 β 3322.79 / 3402.23 Β±67.56 / 3498.82 ms β 3336.20 / 3426.93 Β±71.77 / 3553.38 ms β no change β
β QQuery 23 β 9532.23 / 9717.90 Β±173.93 / 10040.35 ms β 9616.71 / 9829.13 Β±190.87 / 10187.33 ms β no change β
β QQuery 24 β 100.82 / 102.75 Β±1.88 / 105.61 ms β 99.84 / 133.59 Β±60.54 / 254.58 ms β 1.30x slower β
β QQuery 25 β 359.50 / 361.06 Β±2.08 / 365.18 ms β 358.58 / 363.98 Β±9.18 / 382.29 ms β no change β
β QQuery 26 β 100.19 / 101.57 Β±1.83 / 104.85 ms β 100.05 / 101.58 Β±0.99 / 102.56 ms β no change β
β QQuery 27 β 1198.84 / 1231.13 Β±31.86 / 1281.22 ms β 1193.35 / 1213.08 Β±12.16 / 1228.14 ms β no change β
β QQuery 28 β 3622.83 / 3664.37 Β±32.92 / 3718.55 ms β 3670.71 / 3708.65 Β±23.77 / 3737.85 ms β no change β
β QQuery 29 β 40.70 / 59.73 Β±22.76 / 93.23 ms β 40.52 / 40.64 Β±0.12 / 40.85 ms β +1.47x faster β
β QQuery 30 β 574.94 / 587.68 Β±8.96 / 598.55 ms β 574.07 / 588.86 Β±9.27 / 601.17 ms β no change β
β QQuery 31 β 572.41 / 590.13 Β±13.19 / 609.38 ms β 556.12 / 568.65 Β±12.53 / 588.84 ms β no change β
β QQuery 32 β 1128.50 / 1207.71 Β±60.70 / 1283.45 ms β 1100.72 / 1271.47 Β±178.42 / 1599.83 ms β 1.05x slower β
β QQuery 33 β 14819.92 / 15629.35 Β±570.33 / 16473.81 ms β 15108.04 / 15707.30 Β±465.44 / 16419.82 ms β no change β
β QQuery 34 β 15465.47 / 15922.93 Β±360.66 / 16471.78 ms β 15862.02 / 16311.08 Β±384.37 / 16771.79 ms β no change β
β QQuery 35 β 358.01 / 379.98 Β±14.74 / 392.98 ms β 363.71 / 388.46 Β±16.18 / 410.10 ms β no change β
β QQuery 36 β 192.32 / 215.11 Β±23.22 / 253.01 ms β 200.11 / 209.93 Β±8.33 / 219.77 ms β no change β
β QQuery 37 β 98.94 / 105.33 Β±7.33 / 119.34 ms β 100.27 / 101.91 Β±1.00 / 103.16 ms β no change β
β QQuery 38 β 82.47 / 89.30 Β±4.57 / 96.00 ms β 81.10 / 90.57 Β±5.04 / 95.55 ms β no change β
β QQuery 39 β 431.00 / 456.85 Β±19.25 / 484.95 ms β 453.91 / 473.25 Β±20.24 / 512.06 ms β no change β
β QQuery 40 β 15.09 / 25.80 Β±13.03 / 45.20 ms β 16.71 / 17.29 Β±0.60 / 18.34 ms β +1.49x faster β
β QQuery 41 β 14.43 / 16.44 Β±3.14 / 22.67 ms β 14.79 / 15.06 Β±0.15 / 15.25 ms β +1.09x faster β
β QQuery 42 β 14.12 / 14.27 Β±0.21 / 14.69 ms β 14.53 / 14.83 Β±0.26 / 15.26 ms β no change β
βββββββββββββ΄ββββββββββββββββββββββββββββββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββββββββββ΄ββββββββββββββββ
ββββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββ
β Benchmark Summary β β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 82635.20ms β
β Total Time (rich-T-kid_dict-like-preserve) β 83066.15ms β
β Average Time (HEAD) β 1921.75ms β
β Average Time (rich-T-kid_dict-like-preserve) β 1931.77ms β
β Queries Faster β 3 β
β Queries Slower β 4 β
β Queries with No Change β 36 β
β Queries with Failure β 0 β
ββββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββ
```
</p>
</details>
<details><summary>Resource Usage</summary>
**clickbench_partitioned β base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 415.1s |
| Peak memory | 13.0 GiB |
| Avg memory | 6.1 GiB |
| CPU user | 3599.1s |
| CPU sys | 227.8s |
| Peak spill | 0 B |
**clickbench_partitioned β branch**
| Metric | Value |
|--------|-------|
| Wall time | 420.1s |
| Peak memory | 12.9 GiB |
| Avg memory | 6.0 GiB |
| CPU user | 3616.1s |
| CPU sys | 230.5s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org