[
https://issues.apache.org/jira/browse/FLINK-1523?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14385800#comment-14385800
]
ASF GitHub Bot commented on FLINK-1523:
---------------------------------------
Github user andralungu commented on the pull request:
https://github.com/apache/flink/pull/537#issuecomment-87425953
Hi @vasia ,
1). I will add the documentation once we reach a consensus. No use
documenting something that might change :)
2). For the public setter: I still couldn't think of a workaround. Waiting
for suggestions.
3). Most of the comments were addressed in this final commit, however, when
speaking about avoiding unnecessary overhead, the only way I could do that is
to add another VertexCentricIteration class that operates on Tuple3 and have
the user decide from the constructor which class he/she wants to use. To me
that seems to introduce a ton of duplicate code which will look very bad.
Keeping in mind that I need to store the degrees in the vertex value, any
suggestions on how to eliminate this overhead would be very useful.
Thanks!
> Vertex-centric iteration extensions
> -----------------------------------
>
> Key: FLINK-1523
> URL: https://issues.apache.org/jira/browse/FLINK-1523
> Project: Flink
> Issue Type: Improvement
> Components: Gelly
> Reporter: Vasia Kalavri
> Assignee: Andra Lungu
>
> We would like to make the following extensions to the vertex-centric
> iterations of Gelly:
> - allow vertices to access their in/out degrees and the total number of
> vertices of the graph, inside the iteration.
> - allow choosing the neighborhood type (in/out/all) over which to run the
> vertex-centric iteration. Now, the model uses the updates of the in-neighbors
> to calculate state and send messages to out-neighbors. We could add a
> parameter with value "in/out/all" to the {{VertexUpdateFunction}} and
> {{MessagingFunction}}, that would indicate the type of neighborhood.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
