Sounds like this is a tough one. I think doing master compute after vertices is a little counter intuitive but if it fits the existing framework the best, then its the best option. I do think its a priority to make sure the super step numbers a user sees when working with a master compute or vertex in application code should like up perfectly so they see the sequencing as rational, and appearing to run master compute before the vertices even though it actually happens at the end of the previous superstep. If this means inserting a well-commented "X+1" in the plumbing code, it seems worth it to shield the user from having to work with master.compute 2 and vertex superstep 3 as if they were the same superstep.
On Thu, Aug 16, 2012 at 2:15 PM, Maja Kabiljo (JIRA) <j...@apache.org>wrote: > > [ > https://issues.apache.org/jira/browse/GIRAPH-297?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel] > > Maja Kabiljo reopened GIRAPH-297: > --------------------------------- > > > > 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 > > Attachments: GIRAPH-297-1.patch, GIRAPH-297-fix.patch, > GIRAPH-297.patch > > > > > > 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 > > >
