Github user viirya commented on a diff in the pull request:
https://github.com/apache/spark/pull/18301#discussion_r124574957
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/aggregate/TungstenAggregationIterator.scala
---
@@ -367,6 +367,10 @@ class TungstenAggregationIterator(
}
}
+ // Updating average hashmap probe after processing input rows. So even
the iterator of result
+ // rows is not consumed to the last record, we still can show the metric.
+ avgHashProbe.set(hashMap.getAverageProbesPerLookup())
--- End diff --
hmm, the peak memory of hashMap should not be changed during iterating the
rows. But looks like the memory usage of the external sorter will change. It
might be meaningful to update `peakExecutionMemory` at the last record.
Maybe it is also good to use task completion listener to update it?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
