qpid-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andy Goldstein (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (QPID-3280) Message throughput suffers in a cluster when TTL set
Date Thu, 26 May 2011 13:49:47 GMT

    [ https://issues.apache.org/jira/browse/QPID-3280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13039708#comment-13039708
] 

Andy Goldstein commented on QPID-3280:
--------------------------------------

Attached potential patch to resolve issue. Added a new "cluster clock" CPG message that the
cluster elder fires every so often (default is 10ms, but is configurable). Each node stores
and uses the cluster clock when calculating message expiration time or determining if a message
has expired. Also moved the QueueCleaner to the cluster timer when the broker is part of a
cluster.

> Message throughput suffers in a cluster when TTL set
> ----------------------------------------------------
>
>                 Key: QPID-3280
>                 URL: https://issues.apache.org/jira/browse/QPID-3280
>             Project: Qpid
>          Issue Type: Bug
>          Components: C++ Broker
>    Affects Versions: 0.10
>            Reporter: Andy Goldstein
>         Attachments: cluster_ttl.diff
>
>
> When sending a large number of messages with nonzero TTLs to a cluster, overall message
throughput drops by around 20-30% compared to messages with TTL 0.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project:      http://qpid.apache.org
Use/Interact: mailto:dev-subscribe@qpid.apache.org


Mime
View raw message