siknezevic edited a comment on pull request #27246: URL: https://github.com/apache/spark/pull/27246#issuecomment-646931795
> > Could you please let me know would it be OK to hard-code the read buffer size to 1024? > > You think the performance is independent of running platforms, e.g., CPU arch and disk I/O? I'm not 100% sure that the `1024` value is the best on our supported platforms... > > > With 10TB TPCDS data set I tested spilling with query q14a and buffer size of 1024. Execution with hard-coded read buffer size is faster by 37% (27 min vs 37 min) comparing to the execution when buffer size is parameterized and the same size 1024 is used. Query q14a, for 10TB data set, generates around 180 million joins per partition and when buffer size is parameterized, that translates into 10 min longer execution time. > > Why does the parameterized one have so much overhead? Not sure. It looks that call to package.scala to read parameter takes some time. And that time is big enough to cause performance hit because it is executed for each join row. In the case of 10TB data set there is around 180 million rows per partition. The code change that I did was to put 1024 when array is created and comment call to read parameter from package.scala. That is all. ---------------------------------------------------------------- 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: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org