[
https://issues.apache.org/jira/browse/GIRAPH-292?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13432842#comment-13432842
]
Eli Reisman commented on GIRAPH-292:
------------------------------------
During the compute phases the workers have the information about the amount and
size of their owned partitions, and could easily keep count of the vertices
they have processed per superstep and report some kind of percentage progress
via TaskAttemptContext#setStatus() where it will show up on the Hadoop HTML
"Map Details" link. Something like this happens during BspServiceWorker during
reads of input splits every 500k vertices (somewhere around lines 500-550 I
think?)
What sort of output (and where) do you see from the TimedLogger? I had noticed
it in the code but have not noted the output anyplace I've been watching during
job runs.
> Computation progress is not being logged
> ----------------------------------------
>
> Key: GIRAPH-292
> URL: https://issues.apache.org/jira/browse/GIRAPH-292
> Project: Giraph
> Issue Type: Bug
> Reporter: Alessandro Presta
>
> For better understanding long-running jobs, we need a way to log the progress
> of computations during a superstep.
> Looking at GraphMapper, I noticed we have a TimedLogger that's supposed to do
> just that (with a 15 seconds frequency), but I don't see its output.
> Can you guys confirm the issue?
--
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
