[
https://issues.apache.org/jira/browse/HDFS-10645?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15386748#comment-15386748
]
Akira Ajisaka commented on HDFS-10645:
--------------------------------------
Thank you for the update.
# blockReportSizes must be synchronized. I recommend to use
{{Collections.synchronizedSortedSet(new TreeSet<>())}};
# Would you make blockReportSizes final? That way we can avoid null check in
getMaxBlockReportSize.
# Would you use {{!blockReportSizes.isEmpty()}} instead of
{{blockReportSizes.size() > 0}} ?
# In the regression test, would you check the followings?
* the max block report size is greater than zero
* the max data length is equal to 64MB
> Make block report size as a metric and add this metric to datanode web ui
> -------------------------------------------------------------------------
>
> Key: HDFS-10645
> URL: https://issues.apache.org/jira/browse/HDFS-10645
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode, ui
> Reporter: Yuanbo Liu
> Assignee: Yuanbo Liu
> Attachments: HDFS-10645.001.patch, HDFS-10645.002.patch,
> HDFS-10645.003.patch, Selection_047.png, Selection_048.png
>
>
> Record block report size as a metric and show it on datanode UI. It's
> important for administrators to know the bottleneck of block report, and the
> metric is also a good tuning metric.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
