hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steve Loughran (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (MAPREDUCE-6823) FileOutputFormat to support configurable FileOutputCommitter factory
Date Fri, 16 Dec 2016 19:42:00 GMT

     [ https://issues.apache.org/jira/browse/MAPREDUCE-6823?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Steve Loughran updated MAPREDUCE-6823:
    Attachment: HADOOP-13786-HADOOP-13345-001.patch

This is the initial HADOOP-13786 001  PoC patch, to give the MR bit of the code some testing
too. It adds a new factory for FileOutputFormatter to use when creating committers; the default
one returns {{FileOutputCommitter}} instances as normal; a special S3a one in {{hadoop-aws}}
to handle S3a specific operations.

Now, the other way to do this (given the need to keep the s3a code in the s3a module) would
be to allow a notion of a new algorithm, one which relayed to an implementation of an interface.
That would hand a problem not addressed here: how to address subclasses of {{FileOutputFormat}}
with custom subclasses of {{FileOutputCommitter}}, and make it easier to add committers for
other non-FS-destinations, namely the other object stores. 

However, it would be a more significant change to {{FileOutputCommitter}}; I could go that
way, but it'd need support before I started.

> FileOutputFormat to support configurable FileOutputCommitter factory
> --------------------------------------------------------------------
>                 Key: MAPREDUCE-6823
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-6823
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: mrv2
>    Affects Versions: 3.0.0-alpha2
>         Environment: Targeting S3 as the output of work
>            Reporter: Steve Loughran
>            Assignee: Steve Loughran
>         Attachments: HADOOP-13786-HADOOP-13345-001.patch
> In HADOOP-13786 I'm adding a custom subclass for FileOutputFormat, one which can talk
direct to the S3A Filesystem for more efficient operations, better failure modes, and, most
critically, as part of HADOOP-13345, atomic commit of output. The normal committer relies
on directory rename() being atomic for this; for S3 we don't have that luxury.
> To support a custom committer, we need to be able to tell FileOutputFormat (and implicitly,
all subclasses which don't have their own custom committer), to use our new {{S3AOutputCommitter}}.
> I propose: 
> # {{FileOutputFormat}} takes a factory to create committers.
> # The factory to take a URI and {{TaskAttemptContext}} and return a committer
> # the default implementation always returns a {{FileOutputCommitter}}
> # A configuration option allows a new factory to be named
> # An {{S3AOutputCommitterFactory}} to return a  {{FileOutputCommitter}} or new {{S3AOutputCommitter}}
depending upon the URI of the destination.
> Note that MRv1 already supports configurable committers; this is only the V2 API

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: mapreduce-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: mapreduce-issues-help@hadoop.apache.org

View raw message