[
https://issues.apache.org/jira/browse/GIRAPH-469?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13562191#comment-13562191
]
Eli Reisman commented on GIRAPH-469:
------------------------------------
@nitay, this is a ping to ask -- does your +1 still apply? I did change a tiny
bit of the logic in Graph Mapper refactor so that the context being passed into
GraphMapper at each Mapper API call is not passed down fresh to the
GraphTaskManager (all our encapsulated BSP code) but instead is handed once to
GTM at the setup() mapper call, and after that is used from a reference stored
within GTM. I just ran a job today on YARN with this patch built fresh, and it
runs fine. I think its a fairly safe change, but wanted to run it by someone
one more time for approval just because last time you asked I had not made any
real logic changes during the refactor.
If your "+1" from above is still good, I'll commit this and be done with it ;)
> Cleanup GraphMapper
> -------------------
>
> Key: GIRAPH-469
> URL: https://issues.apache.org/jira/browse/GIRAPH-469
> Project: Giraph
> Issue Type: Improvement
> Reporter: Nitay Joffe
> Assignee: Eli Reisman
> Attachments: GIRAPH-469-1-eli-idea.patch, GIRAPH-469-2.patch,
> GIRAPH-469-3.patch, GIRAPH-469-4.patch, GIRAPH-469-5.patch
>
>
> I don't see why we even call a map() method seeing as we are overriding
> run(). We are clearly not particularly "mapreduce-y" so we should make it our
> entry point more clear than a map(). Also I think we should have something
> like a WorkerThread similar to MasterThread and clean up all of this to just
> creare whichever threads the node is assigned roles of.
> Link to review board:
> https://reviews.apache.org/r/8898/
--
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
