Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cassandra.apache.org
Date: Tue, 7 Jan 2014 22:55:50 +0000 (UTC)
From: "Benedict (JIRA)" <jira@apache.org>
To: commits@cassandra.apache.org
Message-ID: <JIRA.12687128.1388729685045.57717.1389135350780@arcas>
In-Reply-To: <JIRA.12687128.1388729685045@arcas>
References: <JIRA.12687128.1388729685045@arcas>
Subject: [jira] [Commented] (CASSANDRA-6544) Reduce GC activity during
 compaction
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/CASSANDRA-6544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13864826#comment-13864826 ] 

Benedict commented on CASSANDRA-6544:
-------------------------------------

Yep, think it's a great idea. Not totally certain if a "slab" allocator or just a single BB per source file that is ~10% larger than the largest row we've visited (with a lower bound of, say, 16K to start with) would do - probably simpler and makes dealing with columns larger than a slab easier, and if we don't share we can safely recycle after every row merge. Will require pulling apart some of the deserializers to support the allocator.


> Reduce GC activity during compaction
> ------------------------------------
>
>                 Key: CASSANDRA-6544
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6544
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Vijay
>            Assignee: Vijay
>             Fix For: 2.1
>
>
> We are noticing increase in P99 while the compactions are running at full stream. Most of it is because of the increased GC activity (followed by full GC).
> The obvious solution/work around is to throttle the compactions, but with SSD's we can get more disk bandwidth for reads and compactions.
> It will be nice to move the compaction object allocations off heap. First thing to do might be create a Offheap Slab allocator with the size as the compaction in memory size and recycle it. 
> Also we might want to make it configurable so folks can disable it when they don't have off-heap memory to reserve.


--
This message was sent by Atlassian JIRA
(v6.1.5#6160)