El Jueves, 18 de Noviembre de 2004 00:59, Jeff Squyres escribió:
> LAM users --
>
> If you could indulge me for a minute, I'd like to take a poll of all
> you "regular users" out there. As you know, we're working heavily on
> Open MPI (http://www.open-mpi.org/). We anticipate a first stable
> release in 1Q 2005.
>
> SHORT VERSION:
> --------------
>
> 1. If Open MPI uses a build system that requires extra tools (such as
> cmake or jam or ...) to be installed in order to be built from source,
> would this be a deterrent to you installing Open MPI from a source
> tarball?
No, it will not be a problem for me. But I guess it can be a problem if
you want Open MPI to go mainstream. I mean, if you have to do the
administration of a linux cluster I prefer to have packages, let's say
rpm's. I do not know if my distribution will have the required packages
but if that is not the case I have to build them up if I want to stay
on a system built with rpm's which is nice for upgrading and so.
That means that for the admin the less the other tools the best.
On the other hand if the Open MPI people have to manage with
several different flavors of the same soft and version for providing
such rpm's I think it is manpower that it is perhaps needed in other
stuff.
>
> 2. If you answered yes to #1, what kind of system will you want to use
> Open MPI on? I.e., what [specific] flavor of system (architecture,
> operating system and version, etc.) would we need to provide a binary
> version of Open MPI for you to install?
>
>
I think that on the cluster side you almost never need a binary. There is
too much fine tuning to do since there are many hardware options and
alternatives. Almost always I prefer to compile :-)
All the best and saludos
--
Jaime D. Perea Duarte. <jaime at iaa dot es>
Linux registered user #10472
Dep. Astrofisica Extragalactica.
Instituto de Astrofisica de Andalucia (CSIC)
Apdo. 3004, 18080 Granada, Spain.
|
