HeartSaVioR commented on PR #50124: URL: https://github.com/apache/spark/pull/50124#issuecomment-2696069749
If you read through my explanation, you would know it's not about the latency slowdown of "every single" row, but maybe a single row per 1000-2000 rows (or even higher, depending on the Kafka poll size). So the microbenchmark you asked only makes sense if you pick "Pmax" rather than "P99" or so. This isn't the "overall" performance improvement. This is a fix for an edge case. I'd also argue that the fix itself is good with the purpose of better code. That's why I asked which one you are not sure. -- 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: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
