[
https://issues.apache.org/jira/browse/HDDS-12919?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sammi Chen resolved HDDS-12919.
-------------------------------
Fix Version/s: 2.1.0
Resolution: Fixed
> Remove redundant FileLock from ChunkWrite
> -----------------------------------------
>
> Key: HDDS-12919
> URL: https://issues.apache.org/jira/browse/HDDS-12919
> Project: Apache Ozone
> Issue Type: Improvement
> Reporter: Sammi Chen
> Assignee: Sammi Chen
> Priority: Major
> Labels: pull-request-available
> Fix For: 2.1.0
>
>
> Currently before chunk read and write, it will acquire a read or write lock
> from fileStripedLock in ChunkUtils, so that read and write are welly
> controlled in Datanode.
> For chunk write, it further acquires an exclusive FileLock after acquire the
> write lock from fileStripedLock. As the FileLock mainly acts as
> inter-process lock purpose, to prevent another process from reading or
> writing to the same lock region, which is not a case in Ozone, there are no
> other processes which will read/write the same block file as Datanode. So
> acquire this exclusive FileLock after already holding the write lock from
> fileStripedLock is not necessary.
> This task aims to remove the FileLock acquire for chunk write.
> BTW, it took ~2ms to acquire a FileLock, ~0.15ms to acquire a LOCK from
> fileStripedLock.
> And FileLock doesn't prevent file get deleted by another thread.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
