[ 
https://issues.apache.org/jira/browse/MAHOUT-1570?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14577908#comment-14577908
 ] 

Dmitriy Lyubimov commented on MAHOUT-1570:
------------------------------------------

I think the code is very nice and streamlines some of by now convoluted history 
of general algebraic operators we did for Spark. 

There are few comments there i have, e.g. AewScalar (i.e. A + 5 kind of stuff) 
is now deprecated in favor of more generic AUnaryFunc which is part of #135. 

#135 is going in this week, so it would be nice to do a few adaptations to some 
changes on logical level and make a PR. without PR it is hard to make 
suggestions. 

If there are any Flink-specific instructions about how to try this out (apart 
from unit tests which i suppose run actual Flink in local mode), it would be 
very welcome to have that added to Mahout website. If there's more than one 
page, we can add a series of menus there.

Thank you for doing this, this is great.

> Adding support for Apache Flink as a backend for the Mahout DSL
> ---------------------------------------------------------------
>
>                 Key: MAHOUT-1570
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-1570
>             Project: Mahout
>          Issue Type: Improvement
>            Reporter: Till Rohrmann
>            Assignee: Suneel Marthi
>              Labels: DSL, flink, scala
>
> With the finalized abstraction of the Mahout DSL plans from the backend 
> operations (MAHOUT-1529), it should be possible to integrate further backends 
> for the Mahout DSL. Apache Flink would be a suitable candidate to act as a 
> good execution backend. 
> With respect to the implementation, the biggest difference between Spark and 
> Flink at the moment is probably the incremental rollout of plans, which is 
> triggered by Spark's actions and which is not supported by Flink yet. 
> However, the Flink community is working on this issue. For the moment, it 
> should be possible to circumvent this problem by writing intermediate results 
> required by an action to HDFS and reading from there.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to