[
https://issues.apache.org/jira/browse/HDFS-5681?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13854085#comment-13854085
]
Kihwal Lee commented on HDFS-5681:
----------------------------------
+1. Because of the synchronizations in LeaseManager, it won't make lease
renewals to be concurrent. However, switching to read locking allows other read
ops to be served. Since fsn write locking is disruptive to NN's parallelism and
the lease renewal is one of the most frequently called write ops, this change
will certainly improve the performance of heavily loaded NNs.
> renewLease should not hold fsn write lock
> -----------------------------------------
>
> Key: HDFS-5681
> URL: https://issues.apache.org/jira/browse/HDFS-5681
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode
> Affects Versions: 2.0.0-alpha, 3.0.0
> Reporter: Daryn Sharp
> Assignee: Daryn Sharp
> Attachments: HDFS-5681.patch
>
>
> Lease renewals unnecessarily hold the write lock which reduces throughput.
> Currently the fsn lock doubles as an "operational lock" to prevent state
> transitions. If the lease manager is thread-safe then a read lock is
> sufficient.
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)
