ankitsultana commented on PR #16084: URL: https://github.com/apache/pinot/pull/16084#issuecomment-2967172806
> // For MSE queries, when this option is set to true, only use servers for leaf stages as the workers for the > // intermediate stages. This is useful to control the fanout of the query and reduce data shuffling. Yeah this is what the new optimizer does by default. Specifically, it only uses the left-most table-scan's servers for intermediate stages. I was meaning to add knobs to tune this behavior (e.g. leverage all servers in a given tenant for intermediate stage), but intended to keep this the default. I'll be raising a PR for that soon. > // TODO: Consider short-circuiting it and directly calculating the result on broker. I think the new query optimizer can do this already. I'll add a E2E test to verify this and address it if it's not already the case. -- 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: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org
