You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by Mukund Madhav Thakur <mt...@cloudera.com.INVALID> on 2022/06/22 20:58:46 UTC

Merged HADOOP-18103 (Vectored IO) with a merge commit in trunk.

Hi Everyone,
Today we merged Vectored IO feature (
https://issues.apache.org/jira/browse/HADOOP-18103)  from
https://github.com/apache/hadoop/tree/feature-vectored-io feature branch
into apache trunk and included a merge commit to the top containing all the
details of child commits.

We will be cherry-picking the individual commits and backport to
branch-3.3.

Me and @Steve Loughran <st...@cloudera.com>  are planning to do a big
3.4.0 release of Hadoop in Q32022 which is going to include the vectored io
feature.

Re: Merged HADOOP-18103 (Vectored IO) with a merge commit in trunk.

Posted by Steve Loughran <st...@cloudera.com.INVALID>.
well done for this

On Wed, 22 Jun 2022 at 21:59, Mukund Madhav Thakur <mt...@cloudera.com>
wrote:

>
> Me and @Steve Loughran <st...@cloudera.com>  are planning to do a big
> 3.4.0 release of Hadoop in Q32022 which is going to include the vectored io
> feature.
>

note that this will be something off branch-3.3 not trunk

he and i were discussing numbering there...i think we've come back to
making it a 3.3.x release (3.3.5?)

why so?

   1. if we gave it a 3.4.0 number it'd (a) confuse people where for a long
   time we have been saying 3.4.0 == trunk
   2. there'd be slower takeup
   3. we would end up having to maintain 3.3.3. with security patches for a
   long time.

there will a fair number of changes in branch 3.3. including s3guard cut,
vectored io, manifest committer...but the core is the same