[GitHub] [arrow] jorisvandenbossche commented on issue #35498: [C++][Parquet] Parquet write_to_dataset performance regression

["jorisvandenbossche (via GitHub)" <gi...@apache.org>]

jorisvandenbossche commented on issue #35498:
URL: https://github.com/apache/arrow/issues/35498#issuecomment-1539766664

   @alexhudspith Thanks a lot for the report and the detailed analysis.
   
   It's unfortunate that this got into the release, as it seems we actually also could see this in our own benchmarks (https://conbench.ursa.dev/benchmark-results/2b587cc1079f4e3a97f542e6f11e883e/, we need some better process to check regressions before a release)
   
   A call to `EnsureAlignment` indeed got added in the PR you reference (the full for-loop got added):
   
   https://github.com/apache/arrow/blob/de6c3cd2b6048641251fac0c64b70c2cd166e0c9/cpp/src/arrow/acero/source_node.cc#L106-L113
   
   I am not directly sure for the reason to do this here, cc @rtpsw @westonpace 
   
   
   


-- 
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@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org