[
https://issues.apache.org/jira/browse/HADOOP-18944?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Steve Loughran updated HADOOP-18944:
------------------------------------
Description:
we are using the wrong default value for the async drain threshold in
{{org.apache.hadoop.fs.s3a.impl.OpenFileSupport}}
{code}
builderSupport.getPositiveLong(ASYNC_DRAIN_THRESHOLD,
defaultReadAhead))
{code}
this means the drain threshold isn't being taken up in random io unless you
have a very small readahead
was:
we are using the wrong default value for the async drain threshold in
{{org.apache.hadoop.fs.s3a.impl.OpenFileSupport))
{code}
builderSupport.getPositiveLong(ASYNC_DRAIN_THRESHOLD,
defaultReadAhead))
{code}
this means the drain threshold isn't being taken up in random io unless you
have a very small readahead
> s3a openfile is using readahead rather than async drain threshold to
> calculate drain threshold
> ----------------------------------------------------------------------------------------------
>
> Key: HADOOP-18944
> URL: https://issues.apache.org/jira/browse/HADOOP-18944
> Project: Hadoop Common
> Issue Type: Bug
> Components: fs/s3
> Affects Versions: 3.3.6
> Reporter: Steve Loughran
> Priority: Minor
>
> we are using the wrong default value for the async drain threshold in
> {{org.apache.hadoop.fs.s3a.impl.OpenFileSupport}}
> {code}
> builderSupport.getPositiveLong(ASYNC_DRAIN_THRESHOLD,
> defaultReadAhead))
> {code}
> this means the drain threshold isn't being taken up in random io unless you
> have a very small readahead
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
