Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hadoop.apache.org
Received-SPF: pass (nike.apache.org: domain of harsh@cloudera.com designates
 209.85.223.176 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAHtRq2fvHdbPZsSPLYxq6QhmfzdKAw5stb=2Dp7Vt=0L5AJb8Q@mail.gmail.com>
References: 
 <CAHtRq2ejyp7i3c24pJK9bwA0gd4=FzNtsayUOHf0o9b3WBUytg@mail.gmail.com>
 <CACBYqwSqn0ZirwyckyWVUG+gbJesR_4bX+WaTmNpYP8Qbp7AyQ@mail.gmail.com>
 <CAHtRq2fvHdbPZsSPLYxq6QhmfzdKAw5stb=2Dp7Vt=0L5AJb8Q@mail.gmail.com>
From: Harsh J <harsh@cloudera.com>
Date: Fri, 19 Oct 2012 11:15:37 +0530
Message-ID: 
 <CAOcnVr2iHct8uBP6U33pdgu3DsyLE9h9Lpkf3O2+dW--p3oHjw@mail.gmail.com>
Subject: Re: Differences between YARN and Hadoop
To: user@hadoop.apache.org
Content-Type: text/plain; charset=ISO-8859-1

First, see what a TLP means, here:
http://incubator.apache.org/guides/graduation.html#introduction.

Apache Hadoop is a collection of 4 major sub-projects (Common, HDFS,
YARN, and MR) presently. The top level directory structure can be
found at the SVN location:
http://svn.apache.org/repos/asf/hadoop/common/trunk/

MRv2 is a sub-project of Apache Hadoop, not a TLP. MRv2 resides under
Hadoop, at http://svn.apache.org/repos/asf/hadoop/common/trunk/hadoop-mapreduce-project/.

YARN is also still a sub-project of Apache Hadoop, not a TLP yet. YARN
resides under Hadoop, at
http://svn.apache.org/repos/asf/hadoop/common/trunk/hadoop-yarn-project/

HDFS is also a sub-project of Apache Hadoop, not a TLP. HDFS resides
under Hadoop, at
http://svn.apache.org/repos/asf/hadoop/common/trunk/hadoop-hdfs-project/

Hope this clears your confusion :)

On Fri, Oct 19, 2012 at 3:58 AM, Tom Brown <tombrown52@gmail.com> wrote:
> Right now Hadoop seems to refer to the bundle of the MRv1 framework
> and HDFS though both can be used independently. Since MRv2 is its own
> top level project, where does leave HDFS?
>
> --Tom
>
> On Thu, Oct 18, 2012 at 5:01 PM, Ruslan Al-Fakikh <metaruslan@gmail.com> wrote:
>> Hi
>> From the beginning YARN was a part of the Hadoop project. Now it has
>> been decided to move it to a top-level apache project.
>> So, now there are two ways of running hadoop jobs:
>> 1) "Old" hadoop (the current stable thing)
>> 2) YARN + MR implementation (the future thing, still not stable AFAIK)
>>
>> Hope that helps
>>
>> On Fri, Oct 19, 2012 at 12:33 AM, Tom Brown <tombrown52@gmail.com> wrote:
>>> To all,
>>>
>>> I have a few questions regarding YARN (with respect to Hadoop):
>>>
>>> Are YARN and Hadoop separate, or is YARN the successor to Hadoop?
>>>
>>> What are the major conceptual differences between YARN and Hadoop?
>>>
>>> I have further questions, but they may or may not make sense depending
>>> on the answers to the above.
>>>
>>> Thanks in advance!
>>>
>>> Tom Brown


-- 
Harsh J