You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@falcon.apache.org by "Shaik Idris Ali (JIRA)" <ji...@apache.org> on 2014/04/22 08:27:18 UTC
[jira] [Comment Edited] (FALCON-246) Upgrade to ActiveMQ 5.9.0
[ https://issues.apache.org/jira/browse/FALCON-246?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13976437#comment-13976437 ]
Shaik Idris Ali edited comment on FALCON-246 at 4/22/14 6:25 AM:
-----------------------------------------------------------------
Hi [~jbonofre],
The patch looks good,
1. the values are hardcoded in log statements, its possible that in future we change the properties without updating the log statements, it might just confuse the users.
{code}
+ kahaDB.setJournalMaxFileLength(32 * 1024 * 1024);
+ kahaDB.setCleanupInterval(2000);
+ kahaDB.setCheckpointInterval(2000);
+ kahaDB.setEnableIndexWriteAsync(true);
+ kahaDB.setConcurrentStoreAndDispatchQueues(true);
+ kahaDB.setConcurrentStoreAndDispatchTopics(true);
+ kahaDB.setIndexCacheSize(2000);
+ kahaDB.setIndexWriteBatchSize(2000);
+ LOGGER.debug(" Persistence adapter: (KahaDB)");
+ LOGGER.debug(" journalMaxFileLength: {}", maxFileLengthVal);
+ LOGGER.debug(" cleanupInterval: 2000");
+ LOGGER.debug(" checkpointInterval: 2000");
+ LOGGER.debug(" enableIndexWriteAsync: true");
+ LOGGER.debug(" concurrentStoreAndDispatchQueues: true");
+ LOGGER.debug(" concurrentStoreAndDispatchTopics: true");
+ LOGGER.debug(" indexWriteBatchSize: 2000");
+ LOGGER.debug(" indexCacheSize: 20000");
{code}
2. it requires rebase.
was (Author: shaik.idris):
Hi [~jbonofre],
The patch looks good,
1. the values are hardcoded in log statements, its possible that in future we change the properties without updating the log statements, it might just confuse the users.
{code}
+ kahaDB.setJournalMaxFileLength(32 * 1024 * 1024);
+ kahaDB.setCleanupInterval(2000);
+ kahaDB.setCheckpointInterval(2000);
+ kahaDB.setEnableIndexWriteAsync(true);
+ kahaDB.setConcurrentStoreAndDispatchQueues(true);
+ kahaDB.setConcurrentStoreAndDispatchTopics(true);
+ kahaDB.setIndexCacheSize(2000);
+ kahaDB.setIndexWriteBatchSize(2000);
+ LOGGER.debug(" Persistence adapter: (KahaDB)");
+ LOGGER.debug(" journalMaxFileLength: {}", maxFileLengthVal);
+ LOGGER.debug(" cleanupInterval: 2000");
+ LOGGER.debug(" checkpointInterval: 2000");
+ LOGGER.debug(" enableIndexWriteAsync: true");
+ LOGGER.debug(" concurrentStoreAndDispatchQueues: true");
+ LOGGER.debug(" concurrentStoreAndDispatchTopics: true");
+ LOGGER.debug(" indexWriteBatchSize: 2000");
+ LOGGER.debug(" indexCacheSize: 20000");
{code}
> Upgrade to ActiveMQ 5.9.0
> -------------------------
>
> Key: FALCON-246
> URL: https://issues.apache.org/jira/browse/FALCON-246
> Project: Falcon
> Issue Type: Improvement
> Components: messaging
> Reporter: Jean-Baptiste Onofré
> Assignee: Jean-Baptiste Onofré
> Attachments: FALCON-246.patch
>
>
> ActiveMQ 5.9.0 brings a lot of improvements, especially in terms of performance.
> I gonna submit a set of patches for this upgrade and leverage the different configurations including performance improvements (kahadb, leveldb, etc) in embedded mode.
--
This message was sent by Atlassian JIRA
(v6.2#6252)