Claudio Martella commented on GIRAPH-45:

Yes, exactly. Given that you'd have to hit the network anyway (so let's ignore 
that cost), the problem is whether the other end would store them on disk or 
not. If so, it can make sense for the sender to hit the disk and then send them 
to the other end at the beginning of next superstep, and that node would have 
to process them as they come. This would mean the disk is hit once on the node 
generating the messages, but then the other end would have to implement a 
pull-model. The other option, the push-model, would still require the second 
end-point to disk the disk twice.
> Improve the way to keep outgoing messages
> -----------------------------------------
>                 Key: GIRAPH-45
>                 URL: https://issues.apache.org/jira/browse/GIRAPH-45
>             Project: Giraph
>          Issue Type: Improvement
>          Components: bsp
>            Reporter: Hyunsik Choi
>            Assignee: Hyunsik Choi
> As discussed in GIRAPH-12(http://goo.gl/CE32U), I think that there is a 
> potential problem to cause out of memory when the rate of message generation 
> is higher than the rate of message flush (or network bandwidth).
> To overcome this problem, we need more eager strategy for message flushing or 
> some approach to spill messages into disk.
> The below link is Dmitriy's suggestion.
> https://issues.apache.org/jira/browse/GIRAPH-12?focusedCommentId=13116253&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13116253

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


Reply via email to