crunch-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Josh Wills (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CRUNCH-218) Add new Target.WriteMode to skip the write and continue pipeline if an output target exists
Date Thu, 13 Jun 2013 19:11:20 GMT

    [ https://issues.apache.org/jira/browse/CRUNCH-218?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13682585#comment-13682585
] 

Josh Wills commented on CRUNCH-218:
-----------------------------------

Hey guys-- so I see two issues here; the first one is also related to something [~jadler]
emailed me about today; i.e., it's hard for the planner to know when it should run a map-only
job w/o some significant hints from the user. The ideal solution to the use case Gabriel outlined
above would be an MSCR fusion that did a pass-through of the map-side output to an identity
reducer, so that we wrote out both outputs on the reduce side and only ran a single job. Obviously,
doing that isn't trivial, otherwise we would have done it already.

I think adding and checking for the _SUCCESS flag (and even checking the timestamps on the
inputs that feed into the checkpointing stuff, as Gabriel suggested) is do-able within the
context of this JIRA. I think the second issue (the one that Gabriel outlines above) should
go into a separate JIRA.
                
> Add new Target.WriteMode to skip the write and continue pipeline if an output target
exists
> -------------------------------------------------------------------------------------------
>
>                 Key: CRUNCH-218
>                 URL: https://issues.apache.org/jira/browse/CRUNCH-218
>             Project: Crunch
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 0.6.0
>            Reporter: Dave Beech
>            Assignee: Josh Wills
>            Priority: Minor
>         Attachments: CRUNCH-218b.patch, CRUNCH-218.patch
>
>
> Quite often I write pipelines which persist data to the filesystem midway through the
process, and then carry on doing further work. 
> If this intermediate data is already present, I think it would be good if I could set
a write mode which skips over this first half of processing. This way I'd avoid running jobs
unnecessarily and wasting cluster resources regenerating data I already have. 
> Example:
> PCollection<B> inter = pipeline.read(source).parallelDo(something).parallelDo(somethingElse);
> inter.write(At.sequenceFile('output'), WriteMode.SKIP_IF_EXISTS);
> PCollection<C> final = inter.parallelDo(moreWork);
>  

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message