Github user pwendell commented on a diff in the pull request:
https://github.com/apache/spark/pull/7770#discussion_r36046965
--- Diff: core/src/main/scala/org/apache/spark/ui/ToolTips.scala ---
@@ -62,6 +62,13 @@ private[spark] object ToolTips {
"""Time that the executor spent paused for Java garbage collection
while the task was
running."""
+ val PEAK_EXECUTION_MEMORY =
+ """Execution memory refers to the memory used by internal data
structures created during
+ shuffles, aggregations and joins when Tungsten is enabled. The
value of this accumulator
+ should be approximately the sum of the peak sizes across all such
data structures created
+ in this task. For SQL jobs, this only tracks all unsafe operators,
broadcast joins, and
+ external sort."""
--- End diff --
The memory is used for other things outside of aggregations, for instance
it is used for the hash join operator and for buffers used during the shuffle.
One thing though, I think we could disable this metric if tungsten is not
turned on. In that case, to your point, we'll be missing out on some other
memory use that isn't currently tracked.
---
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]
