[
https://issues.apache.org/jira/browse/HADOOP-13786?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15755057#comment-15755057
]
Steve Loughran commented on HADOOP-13786:
-----------------------------------------
A few of us (Thomas, Pieter, Mingliang, Aaron and Sean) had a quick conf call
earlier this week, where Thomas and Pieter outlined their proposed algorithm
for implementing zero-rename commits to any consistent S3 endpoint.
I've written up [my interpretation of the
algorithm|https://github.com/steveloughran/hadoop/blob/s3guard/HADOOP-13786-committer/hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/delayed-put-commit.md]
for review, looking though the Hadoop and Spark commit code to see what
appears to be going on, though I do need to actually document the various
algorithms better.
Comments welcome, especially those containing proofs of correctness
> add output committer which uses s3guard for consistent commits to S3
> --------------------------------------------------------------------
>
> Key: HADOOP-13786
> URL: https://issues.apache.org/jira/browse/HADOOP-13786
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3
> Affects Versions: 3.0.0-alpha2
> Reporter: Steve Loughran
> Assignee: Steve Loughran
>
> A goal of this code is "support O(1) commits to S3 repositories in the
> presence of failures". Implement it, including whatever is needed to
> demonstrate the correctness of the algorithm. (that is, assuming that s3guard
> provides a consistent view of the presence/absence of blobs, show that we can
> commit directly).
> I consider ourselves free to expose the blobstore-ness of the s3 output
> streams (ie. not visible until the close()), if we need to use that to allow
> us to abort commit operations.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
