[
https://issues.apache.org/jira/browse/HDFS-16549?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17525376#comment-17525376
]
Mingxiang Li commented on HDFS-16549:
-------------------------------------
in FsDatasetImpl.invalidate the scope protected by block pool write lock is
just do some lightweight operate , it won't take long time when hold lock.If
change to volume lock when user delete batch of blocks, you have to search
volume information in volume map for every block, so I don't think this method
is necessary to change to volume lock
> Consider using volume level lock for deleting blocks
> ----------------------------------------------------
>
> Key: HDFS-16549
> URL: https://issues.apache.org/jira/browse/HDFS-16549
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Yuanbo Liu
> Priority: Major
>
> It's great to see the implement of fine-grain lock for DN has been committed
> into trunk.
> FsDatasetImpl.invalidate is a frequent method to response the delete command
> from NN. How about using volume-level write lock instead of pool-level write
> lock to reduce the cost of write lock.
> cc: [~hexiaoqiao] [~Aiphag0] .
> Thanks for your great work!
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
