-
Notifications
You must be signed in to change notification settings - Fork 588
HDDS-9843. Ozone client high memory (heap) utilization #6153
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
|
will there be a similar effort to improve input stream memory consumption? |
Yes. https://issues.apache.org/jira/browse/HDDS-10283 |
|
This one is blocked by https://issues.apache.org/jira/browse/HDDS-10288. |
jojochuang
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good. It would be nice to enable the leak detection for some of the tests that use BlockOutputStream (e.g. TestBlockOutputStream)
Thanks for the review @jojochuang. All the tests with MiniCluster have LeakDetection enabled by default already. |
|
hmm. leak detected in tests? |
smengcl
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @duongkame . lgtm
btw do we have data from (manual) testing that actually shows the lower client JVM heap usage?
|
Thanks @duongkame for the improvement. Since it is extracted from #5497, I think we should add the following when merging this PR: |
szetszwo
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@duongkame , thanks a lot for working on this!
+1 the change looks good.
|
@jojochuang , @smengcl and @adoroszlai , thanks a lot for reviewing this! |
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1)
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1)
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1) Change-Id: I2f7fd360ef43267ac382e66a1dceb546e06a03c3
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1) Change-Id: Ieaa18864dd4316305df31b05688a5cafa039927c
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1) Change-Id: Ieaa18864dd4316305df31b05688a5cafa039927c
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1) Change-Id: I0cdce3fc27f9a65bcdfd1225dac7bd93ba5ede4f
Co-authored-by: Tsz-Wo Nicholas Sze <[email protected]> (cherry picked from commit 75df6c1) Change-Id: I180c759a4bedb03bf3b83e231dfa1ad17af6425e
What changes were proposed in this pull request?
Ozone client allocates large chunks of heap memory when writing keys (BlockOutputStream, BufferPool).
Extract client-related buffer improvement from #5497.
What is the link to the Apache JIRA
https://issues.apache.org/jira/browse/HDDS-9843
How was this patch tested?
CI: https://github.com/duongkame/ozone/actions/runs/7759444701