[
https://issues.apache.org/jira/browse/HBASE-10156?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13972377#comment-13972377
]
Nicolas Liochon commented on HBASE-10156:
-----------------------------------------
bq. What you seeing? Slow down?
Not yet. I'm tracking the latency peaks on the write path.
bq. If a sync thread came back between your write and just before you go to
call sync
There is still a part that I don't understand here:
Suppose that we have 4 writes by 4 different clients w1, w2, w3, w4. Scenario
would be
w1 gets into the WAL; not sync
w2 gets into the WAL; not sync
w2 finishes: for this it needs to sync, to w1 & are now sync. Client 2 is done
w3 gets into the WAL; not sync
w1 wants to finish. it calls sync, this syncs on w3
that's this last point I'm not sure about? When we sync for w1, do we skip
immediately, or we do sync on w3 because it made it into the WAL in the
meantime?
> FSHLog Refactor (WAS -> Fix up the HBASE-8755 slowdown when low contention)
> ---------------------------------------------------------------------------
>
> Key: HBASE-10156
> URL: https://issues.apache.org/jira/browse/HBASE-10156
> Project: HBase
> Issue Type: Sub-task
> Components: wal
> Reporter: stack
> Assignee: stack
> Fix For: 0.99.0
>
> Attachments: 10156.txt, 10156v10.txt, 10156v11.txt, 10156v12.txt,
> 10156v12.txt, 10156v13.txt, 10156v16.txt, 10156v17.txt, 10156v18.txt,
> 10156v19.txt, 10156v2.txt, 10156v20.txt, 10156v20.txt, 10156v21.txt,
> 10156v21.txt, 10156v21.txt, 10156v3.txt, 10156v4.txt, 10156v5.txt,
> 10156v6.txt, 10156v7.txt, 10156v9.txt, Disrupting.java
>
>
> HBASE-8755 slows our writes when only a few clients. Fix.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
