gratus00 commented on issue #21543: URL: https://github.com/apache/datafusion/issues/21543#issuecomment-4285194351
Hi, related to the "Fix the benchmark" direction, I noticed a small benchmark gap in `datafusion/core/benches/sort.rs`. The existing tuple/string/dictionary cases use `make_sort_exprs(schema)`, which sorts by every column. Would a small benchmark-only PR adding a case that includes a table with a cheap sort-able key such as an `i64` key and non-key utf8/dictionary payload columns but ONLY gets sorted by the `i64` key be useful? The goal would be to make the sort key cheap and better measure the cost of `take` reordering wider payload column cases. I think this case is not covered at the moment. Would love this to be a way I start working on this project! @Dandandan @mbutrovich -- 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]
