hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yu Li (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14969) Add throughput controller for flush
Date Fri, 15 Jan 2016 05:35:39 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14969?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15101226#comment-15101226
] 

Yu Li commented on HBASE-14969:
-------------------------------

{quote}
Does the field have to be volatile ?
I don't see synchronization around access to the field.
{quote}
It was marked as volatile and I didn't change this during the refactoring. I guess this just
follows other Stopable implementation like HRegionServer to make the field volatile. However,
based on current logic, I don't think stop method of ThroughputController will be called in
parallel. I'm hesitating whether to leave it as is or remove the volatile qualifier. What's
your opinion [~tedyu]?
Regarding synchronization, the findbugs warning was in HRegion class not this NoLimitThroughputController,
and I don't think we need synchronization on this volatile since there's no operation like
increment which needs additional atomic

bq. 'is tune' -> 'is tuned'
This is also a copy-paste line... will fix it in the next patch, after got the confirmation
on how to deal with the above volatile filed.

> Add throughput controller for flush
> -----------------------------------
>
>                 Key: HBASE-14969
>                 URL: https://issues.apache.org/jira/browse/HBASE-14969
>             Project: HBase
>          Issue Type: Improvement
>    Affects Versions: 2.0.0
>            Reporter: Yu Li
>            Assignee: Yu Li
>             Fix For: 2.0.0, 1.3.0
>
>         Attachments: HBASE-14969.patch, HBASE-14969_v2.patch, HBASE-14969_v3.patch, HBASE-14969_v4.patch
>
>
> In HBASE-8329 we added a throughput controller for compaction, to avoid spike caused
by huge IO pressure like network/disk overflow. However, even with this control on, we are
still observing disk utils near 100%, and by analysis we think this is caused by flush, especially
when we increase the setting of {{hbase.hstore.flusher.count}}
> In this JIRA, we propose to add throughput control feature for flush, as a supplement
of HBASE-8329 to better control IO pressure.



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

Mime
View raw message