[
https://issues.apache.org/jira/browse/GIRAPH-297?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13433354#comment-13433354
]
Jan van der Lugt commented on GIRAPH-297:
-----------------------------------------
This is expected: the at the end of superstep X, when the master aggregated the
values, it executes master.compute for X+1. This way, it logically executes
_before_ the vertex.computes. Since the master.compute is also executed at the
end of input_superstep, the vertices have the masters values at the beginning
superstep 0. When I implemented this, aggregator values were not checkpointed,
I believe, so I did not think this would interfere with anything. Maybe it's
different now. I agree that this is not the most beautiful approach, but I
tried various and this was the only one I was able to get running (with master
executing before the vertices instead of after).
> Checkpointing on master is done one superstep later
> ---------------------------------------------------
>
> Key: GIRAPH-297
> URL: https://issues.apache.org/jira/browse/GIRAPH-297
> Project: Giraph
> Issue Type: Bug
> Reporter: Maja Kabiljo
> Assignee: Maja Kabiljo
>
> On workers we store checkpoint X before compute() for superstep X are
> executed. On master we do it after those compute() are executed and after
> master.compute() for superstep X+1.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
