Rich-T-kid opened a new pull request, #27339: URL: https://github.com/apache/datafusion/pull/27339
## Which issue does this PR close? - Related to #24822. ## Rationale for this change After a selective filter, a dictionary group key can still carry values that no surviving row references. This experiment compacts those dictionaries before aggregation when the value table is larger than the filtered batch. ## What changes are included in this PR? Adds a physical expression for dictionary compaction and applies it to direct dictionary group keys in partial or single aggregates above filters. ## What is the testing strategy for this PR? Local RLE-dictionary ClickBench results were flat: Q36 moved from 123.72 ms to 123.53 ms, while Q38 did not meet the compaction threshold. Hosted TPC-H, TPC-DS, and partitioned ClickBench results are pending. ## Are there any user-facing changes? No. -- 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]
