[
https://issues.apache.org/jira/browse/GIRAPH-388?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13486367#comment-13486367
]
Maja Kabiljo commented on GIRAPH-388:
-------------------------------------
My mistake, I meant GIRAPH-322, not 314... The changes I have in mind don't try
to improve the case when we have the same message sent to many vertices, as you
are trying to do there (but they don't hurt it either). That's something we
might want to have as an option. Though in a real social graph application, and
running on a large number of workers, the question is how big is the chance
that several destination vertices end up on the same worker.
I did test this with one of our internal applications also, and the speedup was
similar to the one reported. I agree that benchmarks are not good enough to
test all improvements, but in some cases it's a good guideline.
> Improve the way we keep outgoing messages
> -----------------------------------------
>
> Key: GIRAPH-388
> URL: https://issues.apache.org/jira/browse/GIRAPH-388
> Project: Giraph
> Issue Type: Improvement
> Reporter: Maja Kabiljo
> Assignee: Maja Kabiljo
> Attachments: GIRAPH-388.patch
>
>
> As per discussion on GIRAPH-357, in standard application chances that we get
> to use client-side combiner are very low. I experimented with benefits which
> we can get from not having the client-side combiner at all. It turns out that
> having a lot of maps in SendMessageCache, and then collection inside each of
> them, really hurts the performance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
