lidavidm edited a comment on pull request #10557: URL: https://github.com/apache/arrow/pull/10557#issuecomment-875836424
This is quite faster though (actually almost suspiciously so; also, the benchmark loop doesn't include constructing a structarray from individual boolean arrays). Struct: ``` CaseWhenBench32/1048576/0 5780694 ns 5780134 ns 116 bytes_per_second=2.72434G/s CaseWhenBench64/1048576/0 6079444 ns 6078645 ns 115 bytes_per_second=5.16103G/s CaseWhenBench32/1048576/99 5717893 ns 5717306 ns 122 bytes_per_second=2.75402G/s CaseWhenBench64/1048576/99 6160254 ns 6159488 ns 110 bytes_per_second=5.09281G/s CaseWhenBench32Contiguous/1048576/0 1652127 ns 1645047 ns 423 bytes_per_second=7.19786G/s CaseWhenBench64Contiguous/1048576/0 1986278 ns 1986066 ns 334 bytes_per_second=11.8624G/s CaseWhenBench32Contiguous/1048576/99 1826875 ns 1826675 ns 377 bytes_per_second=6.48156G/s CaseWhenBench64Contiguous/1048576/99 2415119 ns 2414697 ns 312 bytes_per_second=9.75582G/s ``` Alternating arguments: (EDIT: updated since I forgot to rebuild first) ``` CaseWhenBench32/1048576/0 32471264 ns 32470765 ns 22 bytes_per_second=496.601M/s CaseWhenBench64/1048576/0 33443346 ns 33443152 ns 21 bytes_per_second=960.585M/s CaseWhenBench32/1048576/99 33171599 ns 33171474 ns 21 bytes_per_second=486.065M/s CaseWhenBench64/1048576/99 33970437 ns 33969940 ns 21 bytes_per_second=945.6M/s CaseWhenBench32Contiguous/1048576/0 7543780 ns 7543809 ns 92 bytes_per_second=1.56961G/s CaseWhenBench64Contiguous/1048576/0 7982627 ns 7982575 ns 86 bytes_per_second=2.95137G/s CaseWhenBench32Contiguous/1048576/99 7894886 ns 7894844 ns 89 bytes_per_second=1.49968G/s CaseWhenBench64Contiguous/1048576/99 8419204 ns 8419033 ns 82 bytes_per_second=2.79811G/s ``` -- 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]
