fanruan-tuco commented on PR #50749:
URL: https://github.com/apache/arrow/pull/50749#issuecomment-5190665330
I investigated the regression in:
FloatRowMajorMatrixConversionFixture<Int64Type>/
ConvertToSparseCSRMatrixInt64
Conbench result:
- Baseline: 998.6 MB/s
- PR #50749: 690.3 MB/s
- Regression: -31.1%
- Equivalent per-iteration time: ~39.8 µs → ~57.6 µs
My local A/B test was run on:
- CPU: Intel Core i9-13900K
- Microarchitecture: Raptor Lake
- 24 cores / 32 threads
- Compiler: GCC 13.3.0
On this system, the benchmark showed no significant regression across
three A/B rounds:
+0.04%, -0.29%, and +0.71%.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
