Rich-T-kid commented on PR #27339: URL: https://github.com/apache/datafusion/pull/27339#issuecomment-6811151441
This tried to compact dictionary group keys after a selective filter so aggregation would not carry and compare dictionary values that no surviving row referenced. We are closing it because that cost was not material in the workloads we tested: local ClickBench Q36 moved only from 123.72 ms to 123.53 ms, Q38 never crossed the activation threshold, TPC-H was entirely unchanged, and partitioned ClickBench was flat overall. TPC-DS had a better total, but the query-level results were mixed rather than a repeatable signal from this path. That is not enough benefit to justify adding a compaction expression and another aggregate planning decision. -- 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]
