Github user JoshRosen commented on the pull request:
https://github.com/apache/spark/pull/5868#issuecomment-99650364
I've been running some manual tests and inspecting the reported shuffle
write metrics as a quick sanity-check to try to understand the causes of this
patch's performance improvements. While doing so, I noticed that the current
sort-shuffle code doesn't report the IO time spent in writing spill files (this
time is actually recorded as part of ShuffleWriteMetrics, but is discarded).
I've filed https://issues.apache.org/jira/browse/SPARK-7413 to start a
discussion of how we should capture and report this time.
---
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]
