[
https://issues.apache.org/jira/browse/HBASE-22387?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16849551#comment-16849551
]
Zheng Hu commented on HBASE-22387:
----------------------------------
I think the periodic QPS dropping in
[QPS-latency.case01.png|https://issues.apache.org/jira/secure/attachment/12970001/QPS-latency.case01.png]
is related to the BlockReaderRemote of HDFS 3.1.0 client, because after the
short-circuit read in my benchmark, the QPS & latency is quite stable now.
I've told this fact to our HDFS/Hadoop team, them will take a look.
> Evaluate the get/scan performance after reading HFile block into offheap
> directly
> ---------------------------------------------------------------------------------
>
> Key: HBASE-22387
> URL: https://issues.apache.org/jira/browse/HBASE-22387
> Project: HBase
> Issue Type: Sub-task
> Reporter: Zheng Hu
> Assignee: Zheng Hu
> Priority: Major
> Attachments: G1GC-stage-stw.case01.png, G1GC-stw.case01.png,
> QPS-latency.case01.png, blockReaderRemote.png, blocksByHFile-stack-trace.png,
> screenshot-1.png, test-cluster-configuration-details.png
>
>
> Now, all sub-tasks has been resolved now (except the HBASE-21946 because of
> the hadoop dependency problem), will provide some performance benckmarks to
> show the latency improvement.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
