andygrove opened a new pull request, #27528: URL: https://github.com/apache/datafusion/pull/27528
## Which issue does this PR close? - Part of #17494. ## Rationale for this change This adds the benchmark from #26447 on its own, as suggested in that PR's review, so the benchmark bot has a baseline on `main` before the hash join change lands. The review of #26447 asked for a skewed build side, where one hot key has many rows and the other keys are unique. With that shape, a probe row that hits the hot key walks a long chain of build rows. None of the existing `hj` queries cover it. ## What changes are included in this PR? Adds Q28 to the `hj` benchmark. The build side has about 6M rows. Line 1 of each order (1.5M rows) shares one key, and every other line has its own key. The probe side has 100K rows, and 100 of them hit the hot key. ## What is the testing strategy for this PR? This PR only adds a benchmark query, so there are no new tests. It runs with `./bench.sh run hj`. ## Are there any user-facing changes? No. This PR was prepared with AI assistance (Claude Code) and reviewed by me. 🤖 Generated with [Claude Code](https://claude.com/claude-code) -- 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]
