tustvold commented on PR #1990: URL: https://github.com/apache/arrow-datafusion/pull/1990#issuecomment-1091446532
Makes sense to me, regardless of what happens with scheduling, having a mechanism to cheaply subdivide the input streams directly, as opposed to streaming the output through a repartitioning operator, seems like a useful feature to have. My expectation is scheduling will help with over-provisioned parallelism in the plan, but will still need mechanisms to express that parallelism in the first place 👍 -- 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]
