Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cassandra.apache.org
Date: Thu, 3 May 2012 20:58:48 +0000 (UTC)
From: "Pavel Yaskevich (JIRA)" <jira@apache.org>
To: commits@cassandra.apache.org
Message-ID: 
 <1698241856.23739.1336078728345.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <1972757074.61307.1327090359799.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] [Commented] (CASSANDRA-3762) AutoSaving KeyCache and System
 load time improvements.
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/CASSANDRA-3762?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13267793#comment-13267793 ] 

Pavel Yaskevich commented on CASSANDRA-3762:
--------------------------------------------

Ok, currently I see following problems in the v3 

  - KeyCacheSerializer.serializingSize(...) method uses AVERAGE_KEY_CACHE_ROW_SIZE where it should use an actual size of the serialized value.
  - in RowCacheTest.java deleted assert should be moved back because we really want to test the amount of read rows properly.
  - AutoSavingCache.loadSaved(...) method should return if exception occurred while loading old style cache.
  - in the AutoSavingCache.loadSaved(...) deleted debug output should be returned because it's very useful for diagnostic purposes.
  - we should probably make CacheSerialize.serialize method to return size of the serialized data, would be useful upon cache writes instead of explicitly calling serializedSize
  - in the ColumnFamilyStore we really should check if we want to load the cache using (caching == Caching.NONE || caching == Caching.ROWS_ONLY). Right now that check if removed and the cache load is called even if there caching was disabled on the given ColumnFamily, the same also applies for the row cache.
  - the following code could be changed (otherwise we would have ifs each time we add a new element):

*From*
{noformat}
if (version == null)
    return new File(conf.saved_caches_directory + File.separator + ksName + "-" + cfName + "-" + cacheType);
else
    return new File(conf.saved_caches_directory + File.separator + ksName + "-" + cfName + "-" + cacheType + "-" + version +".db");
{noformat}

*To*
{noformat}
   return new File(conf.saved_caches_directory + File.separator + ksName + "-" + cfName + "-" + cacheType + "-" + ((version != null) ? version + ".db" : ""));
{noformat}
                
> AutoSaving KeyCache and System load time improvements.
> ------------------------------------------------------
>
>                 Key: CASSANDRA-3762
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-3762
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 1.2
>            Reporter: Vijay
>            Assignee: Vijay
>            Priority: Minor
>             Fix For: 1.2
>
>         Attachments: 0001-CASSANDRA-3762-v2.patch, 0001-CASSANDRA-3762-v3.patch, 0001-SavedKeyCache-load-time-improvements.patch
>
>
> CASSANDRA-2392 saves the index summary to the disk... but when we have saved cache we will still scan through the index to get the data out.
> We might be able to separate this from SSTR.load and let it load the index summary, once all the SST's are loaded we might be able to check the bloomfilter and do a random IO on fewer Index's to populate the KeyCache.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira