emkornfield commented on pull request #7143: URL: https://github.com/apache/arrow/pull/7143#issuecomment-632460492
 OK I think this is ready for review CC @pitrou @wesm Benchmarks show mostly 20% improvement for any random null values across the board. The worse case scenario shows 40-60% regression. This seems to occur when we have deterministically alternating null/not-null. I believe this to be because modern branch predictors can detect this pattern so there is minimal branch cost relative to the extra work that is happening inside of BitRunReader ---------------------------------------------------------------- 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]
