[GitHub] [arrow-datafusion] alamb commented on a diff in pull request #6310: Improve parallelism of repartition operator with multiple cores

["alamb (via GitHub)" <gi...@apache.org>]

alamb commented on code in PR #6310:
URL: https://github.com/apache/arrow-datafusion/pull/6310#discussion_r1188776479


##########
datafusion/core/src/physical_plan/repartition/mod.rs:
##########
@@ -532,9 +541,28 @@ impl RepartitionExec {
                 timer.done();
             }
 
-            // If the input stream is endless, we may spin forever and never yield back to tokio. Hence let us yield.
-            // See https://github.com/apache/arrow-datafusion/issues/5278.
-            tokio::task::yield_now().await;
+            // If the input stream is endless, we may spin forever and

Review Comment:
   I would love some thoughts from reviewers about better heuristics here -- as the comments say I am happy with this heuristic for round robin partitioning but there may be a better way when hash partitioning (like ensure that all channels have at least one batch 🤔 )



-- 
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: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org