You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@activemq.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2018/12/17 14:23:00 UTC
[jira] [Commented] (ARTEMIS-2205) Make AMQP Processing Single
Threaded and other AMQP perf improvements
[ https://issues.apache.org/jira/browse/ARTEMIS-2205?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16723012#comment-16723012 ]
ASF GitHub Bot commented on ARTEMIS-2205:
-----------------------------------------
GitHub user clebertsuconic opened a pull request:
https://github.com/apache/activemq-artemis/pull/2467
ARTEMIS-2205 Performance improvements on AMQP and other parts
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/clebertsuconic/activemq-artemis amqp-PR
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/activemq-artemis/pull/2467.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #2467
----
commit 3199e32d2c98e7cdfd9ef157b9ea0cb19ca85e75
Author: Clebert Suconic <cl...@...>
Date: 2018-12-17T14:11:54Z
ARTEMIS-2205 Refactor AMQP Processing into Single Threaded (per connection)
https://issues.apache.org/jira/browse/ARTEMIS-2205
commit a72c73fad3b8b4ca03737c6733d018dbd385bfbd
Author: Clebert Suconic <cl...@...>
Date: 2018-12-17T14:12:07Z
ARTEMIS-2205 Broker Improvement by caching Routing for most common cases
During AMQP Perf tests this became more relevant on profiling.
It is a general improvement done as part of this AMQP performance task.
commit 6d93d0aff908c23e68a135041d663c3c39701c72
Author: Clebert Suconic <cl...@...>
Date: 2018-12-17T14:12:14Z
ARTEMIS-2205 Avoid new Runnable for every message sent
As we now use the netty executor, creating a new Runnable
for every message received became a relevant CPU, memory and putting extra GC pressure.
By doing this change alone I was able to improve performance by 5% more or less.
https://issues.apache.org/jira/browse/ARTEMIS-2205
commit c11605e00737f570ee4eddc62d9f2bacebcdfb2b
Author: Francesco Nigro <ni...@...>
Date: 2018-12-17T14:12:19Z
ARTEMIS-2205 Optimizing some Lambda usages
https://issues.apache.org/jira/browse/ARTEMIS-2205
----
> Make AMQP Processing Single Threaded and other AMQP perf improvements
> ---------------------------------------------------------------------
>
> Key: ARTEMIS-2205
> URL: https://issues.apache.org/jira/browse/ARTEMIS-2205
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Components: AMQP
> Affects Versions: 2.6.3
> Reporter: clebert suconic
> Assignee: clebert suconic
> Priority: Major
> Fix For: 2.7.0
>
>
> This is using Netty Executor (NIOLoop) to process AMQP.
> Instead of using a lock to enter the AMQP processing, what has issues on scaling up multiple connections per queue, it's using a single threaded executor.
>
>
> This represents a major improvement in AMQP. I will take my time to write a blog post about this with multiple scenarios comparing 2.6.3, master (before this change) and after. (I will post a link here to this JIRA).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)