Re: [gentoo-cluster] Switchable MPI

[Andrew D. Fant]

Donnie Berkholz wrote:
> Andrew D. Fant wrote:
>
>>    I think the real solution here is to use modules (
>>http://modules.sourceforge.net ) to handle the different versions of
>>mpi. I filed a bug about using this for switching alternate virtuals a
>>couple years ago, but nobody seemed to be particularly interested.  I've
>>got a grad student working on something to let users switch between mpi
>>for gcc and mpi for the portland group compilers.  I am most certainly
>>not the right person to do the heavy lifting on this, but I do have an
>>ebuild for modules I could contribute to the cause.
>
>
> That is an idea that's been brought up on the bug, but we need something
> that integrates into the Gentoo framework for switching things around:
> eselect [1].
>
> Perhaps one could create an eselect wrapper around modules. It certainly
> couldn't hurt to take a look at modules (via your ebuild) and see
> whether we could adapt it to our needs.
>
> Thanks,
> Donnie
>
> 1. http://www.gentoo.org/proj/en/eselect/
>

I'll dig around and forward the ebuild to the list later tonight or tomorrow. 
At the risk of going off on a rant, I think that the /etc/env.d model is a core 
part of the problem.   If two users want to use two different mpi 
implementations, will eselect allow them to both use them at the same time, or 
will it have to be one or the other at once?   As I see it, something will have 
to be tweaking $LD_LIBRARY_PATH for each user's environment, or /etc/ld.so.cache 
will have to be changed for each user's login to reflect their prefered mpi 
environment.  $PATH and $MANPATH can be handled by judicious checks in 
/etc/csh.cshrc and /etc/profile, but since /etc/env.d insists on manipulating 
the shared object cache directly, this has the potential to get really ugly 
without some cooperation from the core system team.  I don't think that the 
cluster herd can fix it on it's own.

I would love to be proven wrong and will gladly defer to anyone who does so,  I 
just know that I've been trying to outthink this problem for a year or so on our 
cluster and haven't come up with a sane solution for a true multi-user environment.

Andy

-- 
Andrew Fant           | The lion and the calf shall lie   | Disclaimer:
andrew.fant@×××××.edu | down together, but the calf won't | Do you REALLY
TCCS/USG              | get much sleep.                   | think I can
Tufts University      |                  W. Allen         | speak for Tufts?
-- 
gentoo-cluster@g.o mailing list