You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@hadoop.apache.org by gm...@cs.ucf.edu on 2011/05/02 16:55:33 UTC
Re: questions about hadoop map reduce and compute intensive related applications
> Integrating MPI with map-reduce is currently difficult and/or very ugly,
> however. Not impossible and there are hackish ways to do the job, but
> they are hacks.
There is an project out of Sandia National Lab that puts MR and MPI
together in a library if you're interested -->
http://www.sandia.gov/~sjplimp/mapreduce.html
The project isn't mature yet, and I haven't actually used it myself.
--
--
Grant Mackey
PhD student Computer Engineering
University of Central Florida
Re: questions about hadoop map reduce and compute intensive related applications
Posted by elton sky <el...@gmail.com>.
thanks gmackey,
There is an project out of Sandia National Lab that puts MR and MPI together
> in a library if you're interested -->
> http://www.sandia.gov/~sjplimp/mapreduce.html
That is a implementation of MR using MPI. I saw that as well but haven't
tried it out.
I am actually looking at programming model level integration. You know, for
some applications, you can take advantage of high through put from MR and
message passing from MPI.
-Elton
On Tue, May 3, 2011 at 12:55 AM, <gm...@cs.ucf.edu> wrote:
>
> Integrating MPI with map-reduce is currently difficult and/or very ugly,
>> however. Not impossible and there are hackish ways to do the job, but they
>> are hacks.
>>
>
> There is an project out of Sandia National Lab that puts MR and MPI
> together in a library if you're interested -->
> http://www.sandia.gov/~sjplimp/mapreduce.html
>
> The project isn't mature yet, and I haven't actually used it myself.
>
> --
> --
> Grant Mackey
> PhD student Computer Engineering
> University of Central Florida
>