beam-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <>
Subject [jira] [Commented] (BEAM-22) DirectPipelineRunner: support for unbounded collections
Date Thu, 31 Mar 2016 20:33:25 GMT


ASF GitHub Bot commented on BEAM-22:

GitHub user tgroh opened a pull request:

    [BEAM-22] Clean up InProcess Read Evaluators

    These are a couple of minor improvements to BoundedReadEvaluator
    and UnboundedReadEvaluator that enables splitting a source at
    evaluation time, as well as minor code cleanup.

You can merge this pull request into a Git repository by running:

    $ git pull ippr_cleaner_read_evaluators

Alternatively you can review and apply these changes as the patch at:

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #106
commit b83328c5b763c1974ad5a84d2870753e2078d1ee
Author: Thomas Groh <>
Date:   2016-03-31T17:40:37Z

    Explicitly track the Source a ReadEvaluator is using
    This permits use of sources that are not the initial source used in the
    transform. BoundedSource#splitIntoBundles and
    UnboundedSource#generateInitialSplits generate multiple source objects
    for the same transform in order to permit parallelism.

commit 2f756cc06967afa6d49aae54296568e70145551d
Author: Thomas Groh <>
Date:   2016-03-31T17:43:56Z

    Use proper scoping, interfaces in BoundedReadEvaluator
    Use BoundedReader instead of Reader.
    contentsRemaining should be method-scoped not instance-scoped.


> DirectPipelineRunner: support for unbounded collections
> -------------------------------------------------------
>                 Key: BEAM-22
>                 URL:
>             Project: Beam
>          Issue Type: Improvement
>          Components: runner-direct
>            Reporter: Davor Bonaci
>            Assignee: Thomas Groh
> DirectPipelineRunner currently runs over bounded PCollections only, and implements only
a portion of the Beam Model.
> We should improve it to faithfully implement the full Beam Model, such as add ability
to run over unbounded PCollections, and better resemble execution model in a distributed system.
> This further enables features such as a testing source which may simulate late data and
test triggers in the pipeline. Finally, we may want to expose an option to select between
"debug" (single threaded), "chaos monkey" (test as many model requirements as possible), and
"performance" (multi-threaded).
> more testing (chaos monkey) 
> Once this is done, we should update this StackOverflow question:

This message was sent by Atlassian JIRA

View raw message