avantgardnerio commented on issue #2299: URL: https://github.com/apache/datafusion-ballista/issues/2299#issuecomment-5295752342
I would like to finish up work on parallel windows, and also dive into benchmarks to see which if any could be improved given the RangeRepartitioning primitives now in place. Particularly anything with: 1. a plan where Spark goes faster & uses range partitioning 2. any plan with "hot keys" in hash partitioning - low cardinality or skewed data These would be obvious candidates for improvements by switching to range re-partitioning, though there might be others. I'd also like to continue helping with manageability & HA, as well as general performance improvements. (it's always fun to see a benchmark go faster :) -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
