hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Abhishek Singh Chouhan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14352) Replication is terribly slow with WAL compression
Date Tue, 01 Sep 2015 11:40:46 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14352?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14725243#comment-14725243
] 

Abhishek Singh Chouhan commented on HBASE-14352:
------------------------------------------------

The graph is of two back to back workloads and shows point in time size of log queue of a
particular host(each coloured line is a different host). When replication is complete, it
goes back to 0.For the earlier load you can see the size of log queue going up and then coming
down,at the point it comes back to 0, we have completely replicated all the edits. For the
latter workload you can see that it is very spread out (it took too much time for the queue
to clear up).

> Replication is terribly slow with WAL compression
> -------------------------------------------------
>
>                 Key: HBASE-14352
>                 URL: https://issues.apache.org/jira/browse/HBASE-14352
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 0.98.13
>            Reporter: Abhishek Singh Chouhan
>         Attachments: age_of_last_shipped.png, size_of_log_queue.png
>
>
> For the same load, replication with WAL compression enabled is almost 6x slower than
with compression turned off. Age of last shipped operation is also correspondingly much higher
when compression is turned on. 
> By observing Size of log queue we can see that it is taking too much time for the queue
to clear up.
> Attaching corresponding graphs.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message