Mailing-List: contact dev-help@jackrabbit.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@jackrabbit.apache.org
Message-ID: <31484264.1179326538194.JavaMail.jira@brutus>
Date: Wed, 16 May 2007 07:42:18 -0700 (PDT)
From: "Stefan Guggisberg (JIRA)" <jira@apache.org>
To: dev@jackrabbit.apache.org
Subject: [jira] Commented: (JCR-314) Fine grained locking in
 SharedItemStateManager
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/JCR-314?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12496317 ] 

Stefan Guggisberg commented on JCR-314:
---------------------------------------

> - Accept the patch ISMLocking.patch and discuss a locking implementation/strategy that's better suited than the current one.

+1

btw, i did a couple of test runs using ConcurrentReadWriteTest against
 a) current trunk
 b) a) with ISMLocking.patch applied (using DefaultISMLocking)
 c) b) with FineGrainedISMLocking.patch applied (using FineGrainedISMLocking)

environment: macbook pro (intel, 2ghz), os-x 10.4.9, jre 1.4.2

results: 

a)
**
   ~2'300 #writes
   ~2'300'000 #reads

b)
**
   ~2'300 #writes
   ~2'300'000 #reads

c)
**
   ~1'700 #writes
   ~2'300'000 #reads

=> c) shows a significant decrease in #writes while #reads doesn't seem
to be affected. i guess that FineGrainedISMLocking still has room for 
improvement ;)


> Fine grained locking in SharedItemStateManager
> ----------------------------------------------
>
>                 Key: JCR-314
>                 URL: https://issues.apache.org/jira/browse/JCR-314
>             Project: Jackrabbit
>          Issue Type: Improvement
>          Components: core
>    Affects Versions: 0.9, 1.0, 1.0.1, 1.1, 1.1.1, 1.2.1, 1.2.2, 1.2.3
>            Reporter: Marcel Reutegger
>         Attachments: FineGrainedISMLocking.patch, ISMLocking.patch
>
>
> The SharedItemStateManager (SISM) currently uses a simple read-write lock to ensure data consistency. Store operations to the PersistenceManager (PM) are effectively serialized.
> We should think about more sophisticated locking to allow concurrent writes on the PM.
> One possible approach:
> If a transaction is currently storing data in a PM a second transaction may check if the set of changes does not intersect with the first transaction. If that is the case it can safely store its data in the PM.
> This fine grained locking must also be respected when reading from the SISM. A read request for an item that is currently being stored must be blocked until the store is finished.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.