[
https://issues.apache.org/jira/browse/HDFS-14366?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16791077#comment-16791077
]
Chao Sun commented on HDFS-14366:
---------------------------------
Thanks [~jojochuang]! Yes similar to HDFS-14171, the issue is again caused by
{{getNumLiveDataNodes}}. It is much worse in the {{append}} case though since
the method is called on the hot path and directly affect NameNode performance,
as it holds the write lock while doing so. Here, the proposed change is pretty
simple and IMO risk-free. It'd be great if you can take a look. :)
> Improve HDFS append performance
> -------------------------------
>
> Key: HDFS-14366
> URL: https://issues.apache.org/jira/browse/HDFS-14366
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: hdfs
> Affects Versions: 2.8.2
> Reporter: Chao Sun
> Assignee: Chao Sun
> Priority: Major
> Attachments: HDFS-14366.000.patch, append-flamegraph.png
>
>
> In our HDFS cluster we observed that {{append}} operation can take as much as
> 10X write lock time than other write operations. By collecting flamegraph on
> the namenode (see attachment: append-flamegraph.png), we found that most of
> the append call is spent on {{getNumLiveDataNodes()}}:
> {code}
> /** @return the number of live datanodes. */
> public int getNumLiveDataNodes() {
> int numLive = 0;
> synchronized (this) {
> for(DatanodeDescriptor dn : datanodeMap.values()) {
> if (!isDatanodeDead(dn) ) {
> numLive++;
> }
> }
> }
> return numLive;
> }
> {code}
> this method synchronizes on the {{DatanodeManager}} which is particularly
> expensive in large clusters since {{datanodeMap}} is being modified in many
> places such as processing DN heartbeats.
> For {{append}} operation, {{getNumLiveDataNodes()}} is invoked in
> {{isSufficientlyReplicated}}:
> {code}
> /**
> * Check if a block is replicated to at least the minimum replication.
> */
> public boolean isSufficientlyReplicated(BlockInfo b) {
> // Compare against the lesser of the minReplication and number of live
> DNs.
> final int replication =
> Math.min(minReplication, getDatanodeManager().getNumLiveDataNodes());
> return countNodes(b).liveReplicas() >= replication;
> }
> {code}
> The way that the {{replication}} is calculated is not very optimal, as it
> will call {{getNumLiveDataNodes()}} _every time_ even though usually
> {{minReplication}} is much smaller than the latter.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
