[
https://issues.apache.org/jira/browse/HDFS-9412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15000269#comment-15000269
]
He Tianyi commented on HDFS-9412:
---------------------------------
A simpler approach would be ignore small blocks during iteration (just as
Balancer did in HDFS-8824, if HDFS-8824 is implemented, Balancer don't need
these blocks anyway), this will reduce the cost of gathering enough amount of
size of data (I did not do the profiling but perhaps there are enough overhead
in {{addBlock}}, lots of objects are created).
> getBlocks occupies FSLock and takes too long to complete
> --------------------------------------------------------
>
> Key: HDFS-9412
> URL: https://issues.apache.org/jira/browse/HDFS-9412
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: He Tianyi
> Assignee: He Tianyi
>
> {{getBlocks}} in {{NameNodeRpcServer}} acquires a read lock then may take a
> long time to complete (probably several seconds, if number of blocks are too
> much).
> During this period, other threads attempting to acquire write lock will wait.
> In an extreme case, RPC handlers are occupied by one reader thread calling
> {{getBlocks}} and all other threads waiting for write lock, rpc server acts
> like hung. Unfortunately, this tends to happen in heavy loaded cluster, since
> read operations come and go fast (they do not need to wait), leaving write
> operations waiting.
> Looks like we can optimize this thing like DN block report did in past, by
> splitting the operation into smaller sub operations, and let other threads do
> their work between each sub operation. The whole result is returned at once,
> though (one thing different from DN block report).
> I am not sure whether this will work. Any better idea?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
