alamb commented on PR #7513: URL: https://github.com/apache/arrow-rs/pull/7513#issuecomment-2919166080
My analysis of benchmark results is: 1. Many of the filter kernel benchmarks are dominated by the time required to construct the final Array (as the actual filtering operation is so fast) so adding a Builder in that path slows them down. I am not sure how important this would be in an actual system where filtering is almost followed by coalesce. 2. The reason StringView is so much slower is that I have it automatically coalescing values, which is important for actual systems but clearly a major overhead in the microbenchmarks -- 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-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
