[
https://issues.apache.org/jira/browse/CRUNCH-320?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13864701#comment-13864701
]
Josh Wills commented on CRUNCH-320:
-----------------------------------
No, materialized collections are stored on disk and are streamed into the
client one record at a time. If you never call iterator() on the returned
Iterable<T>, it won't be read into the client at all.
I expect that my patch will fix the problem, yes-- please feel free to apply it
and let me know if there's a problem.
> Materialize several PObject & PCollection objects in parallel (deferred
> materialization)
> ----------------------------------------------------------------------------------------
>
> Key: CRUNCH-320
> URL: https://issues.apache.org/jira/browse/CRUNCH-320
> Project: Crunch
> Issue Type: Improvement
> Components: Core
> Reporter: Jason Gauci
> Assignee: Josh Wills
> Attachments: CRUNCH-320.patch
>
>
> Currently, Crunch blocks and materializes PCollections (through
> foo.materialize()) and PObjects (through foo.getValue()) on demand, but it
> would be a significant performance improvement if we could mark several of
> these objects as to be materialized, and then materialize all of them in
> parallel as part of a pipeline.run() call.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
