[
https://issues.apache.org/jira/browse/FLINK-1885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14499949#comment-14499949
]
ASF GitHub Bot commented on FLINK-1885:
---------------------------------------
Github user vasia commented on the pull request:
https://github.com/apache/flink/pull/598#issuecomment-94002207
Hi @markus-h!
I see the point in having a bulk iteration in Gelly, however I'm not sure I
would add it as a "mode" in vertex-centric iteration. VertexCentricIteration
implements the Pregel model and it might be confusing to change its semantics
like this (or maybe not).
Also, I am not sure whether the messaging-vertexUpdate abstraction is what
you would like to have in a bulk graph iteration.
It might be better to add a bulk graph iteration, where the gathering of
neighborhoods is abstracted and the user only provides the step function, i.e.
something like the neighborhood methods, but iterative.
What do you think?
In any case, I think that a use-case / example would really help motivate
adding this :-)
> Bulk mode for gelly
> -------------------
>
> Key: FLINK-1885
> URL: https://issues.apache.org/jira/browse/FLINK-1885
> Project: Flink
> Issue Type: Improvement
> Components: Gelly
> Reporter: Markus Holzemer
> Assignee: Markus Holzemer
> Priority: Minor
>
> For my research I need to execute graph algorithm with delta iterations and
> with bulk iterations. So I added an execution mode to gelly that executes the
> vertex centric iteration with a bulk iteration.
> Since gelly normally keeps the whole graph in the solution set of the delta
> iteration, it can only handle graphs that fit into the distributed memory of
> ones nodes (correct me if I am wrong). So a bulk mode could also be usefull
> for handling graphs that do not fit in memory.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
