tephra-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lars Hofhansl (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (TEPHRA-299) Executing a large batch delete is very slow
Date Sun, 14 Apr 2019 04:08:00 GMT

    [ https://issues.apache.org/jira/browse/TEPHRA-299?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16817178#comment-16817178
] 

Lars Hofhansl commented on TEPHRA-299:
--------------------------------------

-v2 is better. Directly creates the Puts. Also optimizes the existing code by not getting
the TableDescriptor for each Delete (which is an RPC to the HMaster).

Please have a look.

> Executing a large batch delete is very slow
> -------------------------------------------
>
>                 Key: TEPHRA-299
>                 URL: https://issues.apache.org/jira/browse/TEPHRA-299
>             Project: Tephra
>          Issue Type: Bug
>    Affects Versions: 0.15.0-incubating
>            Reporter: Lars Hofhansl
>            Assignee: Poorna Chandra
>            Priority: Major
>         Attachments: 299-DOES-NOT_WORK.txt, 299-client-v2.txt, 299-client.txt
>
>
> I noticed that batch deletes are quire slow. In the profiler I found that almost all
of the time is spent in org.apache.hadoop.hbase.regionserver.wal.FSHLog.blockOnSync().
> Looking at TransactionProcessor.preDelete it is obvious why:
> The batch delete is translated into *single* puts that are added to the region one by
one, so each time the WAL is flushed.
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message