Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Date: Fri, 6 Mar 2015 19:45:39 +0000 (UTC)
From: "stack (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Message-ID: <JIRA.12776620.1424500574000.23653.1425671139015@Atlassian.JIRA>
In-Reply-To: <JIRA.12776620.1424500574000@Atlassian.JIRA>
References: <JIRA.12776620.1424500574000@Atlassian.JIRA>
 <JIRA.12776620.1424500574747@arcas>
Subject: [jira] [Commented] (HBASE-13082) Coarsen StoreScanner locks to
 RegionScanner
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HBASE-13082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14350787#comment-14350787 ] 

stack commented on HBASE-13082:
-------------------------------

bq. The only reason why I am still hesitant on this one is that we'll now be forced to keep the RegionScanner lock in the future. 

+1. Commit. Go for it. Undoing the locks is a good thing (tm). It cleans up old lazy thinking and shows the way forward.

That the lock is too coarse, we can work on later especially after flush-by-store gets done; then flush and compaction can be made purely Store-scoped.

> Coarsen StoreScanner locks to RegionScanner
> -------------------------------------------
>
>                 Key: HBASE-13082
>                 URL: https://issues.apache.org/jira/browse/HBASE-13082
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>         Attachments: 13082-test.txt, 13082.txt, 13082.txt, gc.png, gc.png, gc.png, hits.png, next.png, next.png
>
>
> Continuing where HBASE-10015 left of.
> We can avoid locking (and memory fencing) inside StoreScanner by deferring to the lock already held by the RegionScanner.
> In tests this shows quite a scan improvement and reduced CPU (the fences make the cores wait for memory fetches).
> There are some drawbacks too:
> * All calls to RegionScanner need to be remain synchronized
> * Implementors of coprocessors need to be diligent in following the locking contract. For example Phoenix does not lock RegionScanner.nextRaw() and required in the documentation (not picking on Phoenix, this one is my fault as I told them it's OK)
> * possible starving of flushes and compaction with heavy read load. RegionScanner operations would keep getting the locks and the flushes/compactions would not be able finalize the set of files.
> I'll have a patch soon.


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