[
https://issues.apache.org/jira/browse/HDDS-8865?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17733354#comment-17733354
]
ChenXi edited comment on HDDS-8865 at 6/16/23 7:11 AM:
-------------------------------------------------------
For the DN Delete command processing
After analysis, we found that DeleteBlocksCommandHandler thread often fails to
obtain Container's write lock due to `reblancer`, `replication`,
`BlockDeletingService` and other threads holding Container locks, We observe
that sometimes, DeleteBlocksCommandHandler will be stuck for hours, but most of
the time, DeleteBlocksCommandHandler will only be stuck for a few minutes, when
SCM will send a lot of invalid transactions to the current blocked DN, and
other DNs will receive fewer transactions.
The Metrics of DeleteBlocksCommandHandler threadpool queue size
!image-2023-06-16-15-09-59-163.png|width=747,height=267!
was (Author: JIRAUSER294158):
For the DN Delete command processing
After analysis, we found that DeleteBlocksCommandHandler thread often fails to
obtain Container's write lock due to `reblancer`, `replication`,
`BlockDeletingService` and other threads holding Container locks, We observe
that sometimes, DeleteBlocksCommandHandler will be stuck for hours, but most of
the time, DeleteBlocksCommandHandler will only be stuck for a few minutes, when
SCM will send a lot of invalid transactions to the current blocked DN, and
other DNs will receive fewer transactions.
The Metrics of DeleteBlocksCommandHandler threadpool queue size
!image-2023-06-16-15-09-59-163.png!
> Ozone asynchronous deletion performance optimization
> ----------------------------------------------------
>
> Key: HDDS-8865
> URL: https://issues.apache.org/jira/browse/HDDS-8865
> Project: Apache Ozone
> Issue Type: Improvement
> Reporter: ChenXi
> Assignee: ChenXi
> Priority: Major
> Attachments: image-2023-06-16-15-09-59-163.png
>
>
> Background:
> We have a Cluster which will write many small key. The user will merge these
> small keys into one big key, so these small keys will not be stored for a
> long time. So we have almost the same QPS of delete requests as writes.
> The current key deletion performance cannot meet the requirements, the disk
> is using less and less available capacity, but in fact, only a small portion
> of the user's valid data
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
