flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexander Alexandrov (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-986) Add intermediate results to distributed runtime
Date Mon, 08 Sep 2014 14:04:28 GMT

    [ https://issues.apache.org/jira/browse/FLINK-986?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125507#comment-14125507
] 

Alexander Alexandrov commented on FLINK-986:
--------------------------------------------

I didn't think about recovery. I guess eager application of updates makes any kind of REDO-based
recovery impossible. 

I wonder know what this means for the current state of the bulk and delta iterations, which
as far as I know are realized in that way (work- and solution set are indexed by key and updated
pointwise in each iteration). How does the current recovery logic cope with that?

> Add intermediate results to distributed runtime
> -----------------------------------------------
>
>                 Key: FLINK-986
>                 URL: https://issues.apache.org/jira/browse/FLINK-986
>             Project: Flink
>          Issue Type: New Feature
>          Components: Distributed Runtime
>            Reporter: Ufuk Celebi
>            Assignee: Ufuk Celebi
>            Priority: Blocker
>
> Support for intermediate results in the runtime is currently blocking different efforts
like fault tolerance or result collection at the client.



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

Mime
View raw message