Jackie-Jiang commented on PR #12354: URL: https://github.com/apache/pinot/pull/12354#issuecomment-1960600876
The result is a little bit counter intuitive. In most cases, when null handling is enabled, it should be much slower than when it is disabled because we need to pay overhead of iterating over the null bitmap, which should be much more expensive than simply summing up all values. Is it possible that other part of the query execution is dominating the results, or some of the execution is skipped? Micro-benchmark is usually not accurate when too much code is tested (especially multi-threaded). Can you set up a benchmark to only test the aggregation function itself? We may construct a `BlockValSet` and pass it into the function. After the execution, we can use a blackhole to consume the value. -- 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: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org