Dandandan commented on PR #19694: URL: https://github.com/apache/datafusion/pull/19694#issuecomment-3892631330
> @Dandandan care to shed some light on the current state of this test? I see you've: > > * Made the repartition threshold smaller > * Disabled join dynamic filter pushdown > * Cherry picked the dynamic row filter pushdown > * Turned on filter pushdown by default > > What's the combined impact of these changes? I'm having trouble tracking the benchmark results AFAIK the non-regression of tpch is a combination of * Cherry picked the dynamic row filter pushdown (your changes) * Disabled join dynamic filter pushdown (the config change) I triggered a new run to verify this in CI -- 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]
