[
https://issues.apache.org/jira/browse/HADOOP-18791?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17742667#comment-17742667
]
Viraj Jasani commented on HADOOP-18791:
---------------------------------------
sounds good, i just realized unbuffer is already in-progress
[https://github.com/apache/hadoop/pull/5832]
> S3A prefetching: switch to prefetching for chosen read policies
> ---------------------------------------------------------------
>
> Key: HADOOP-18791
> URL: https://issues.apache.org/jira/browse/HADOOP-18791
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3
> Affects Versions: 3.3.6
> Reporter: Steve Loughran
> Priority: Major
>
> before switching to prefetching input stream everywhere, add an option to
> list which of the fs.option.openfile.read.policy policies to switch too, e.g
>
> fs.s3a.inputstream.prefetch.policies=whole-file, sequential, adaptive
> this would leave random and vectored on s3a input stream.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
