gsmiller commented on PR #12417: URL: https://github.com/apache/lucene/pull/12417#issuecomment-1686776595
@ChrisHegarty I was considering some experimentation with [vectorized prefix sum implementations](https://en.algorithmica.org/hpc/algorithms/prefix/), but saw your comment above stating: > What bothers me even more is that we cannot easily integrate the prefix sum calculation into the unpack - as we run into Panama bounds check issues that make the performance very poor. I also came across some [benchmarks](https://github.com/jpountz/vectorized-prefix-sum) it looks like you may have collaborated on with @jpountz related to some different prefix sum SIMD approaches. Can you elaborate any more on the performance issues related to these vectorized attempts? I assume the benchmark results were poor for you as well (I've tested on a few different machines with pretty horrid results, but I don't really understand why they're so bad). -- 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: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
