Github user mallman commented on the issue:
https://github.com/apache/spark/pull/14617
Hi @jerryshao. I think this is a great idea and fills in an important gap
in the app's UI. Going by the screenshot you posted, instead of putting both on
and off heap memory in a single column, how about putting them in separate
columns? For one thing, this would make them independently sortable. Also, how
about splitting on- and off-heap memory in the summary table as well?
Another missing piece of the puzzle is the total executor memory that
executors report back to the master. For example, in the standalone master UI
each app only reports allocated heap memory per executor. This is perhaps an
even bigger omission as it misrepresents the actual memory footprint of the
executors, and because of that the scheduler does not compute or report memory
usage per worker correctly. This breaks memory resource allocation. Is this
issue something your patch addresses?
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
