[
https://issues.apache.org/jira/browse/GIRAPH-515?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13578566#comment-13578566
]
Alessandro Presta commented on GIRAPH-515:
------------------------------------------
PageRank on a real-world graph (i.e. a dataset loaded from Hive) was using
around 40GB on each worker.
But that's probably more than it needs, because it didn't do very much GC.
I haven't done a thorough analysis but I imagine we can squeeze it down even
further with this optimization.
> More efficient and flexible edge-based input
> --------------------------------------------
>
> Key: GIRAPH-515
> URL: https://issues.apache.org/jira/browse/GIRAPH-515
> Project: Giraph
> Issue Type: Improvement
> Reporter: Alessandro Presta
> Assignee: Alessandro Presta
> Attachments: GIRAPH-515.patch
>
>
> The current implementation of edge-based input using mutations is not as
> memory-efficient as it could be, and also can't be used with immutable vertex
> classes.
> By having an ad-hoc code path for edges sent during input superstep, we can
> achieve both memory efficiency and eliminate the current restriction.
--
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
