pitrou opened a new pull request #8890: URL: https://github.com/apache/arrow/pull/8890
Add two RecordBatch sorting implementations: * A single-pass left-to-right radix sort that's fast up to ~8 sort keys * A single-pass multiple-key-comparing sort that gives decent performance for large numbers of sort keys Both implementations benefit from direct indexed access into the contiguous RecordBatch columns (as opposed to table sorting, which must index into the chunks). Add some RecordBatch-sorting benchmarks. Also, add and improve tests; and fix a bug related to sorting of NaNs and nulls. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
