[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: PetSc

Adam C Powell IV wrote:
> Also, I could not find any clear indication on this list that lam was so vastly
> superior to mpich.  I would love to hear your (plural, the list's) thoughts and
> experiences.

As a message passing system, the lam implementation seemed to us more
robust and nicely integrated with POSIX systems. AFAIK the mpich is a
bit older, we used to code with it in our distributed systems course 2
1/2 years ago. LAM is easier to use, etc.. And in the performance tests,
it did as well as arch-specific code on IBM Cognitive Computer on our
research cluster.

For beowulf apps, I'd suggest that using LAM is a lot nicer. The older
environments are always usable, but you will for instance find few
projects that use pvm on their new beowulf code.

However, I am not saying in any way that lam outperforms or surpasses
mpich in many ways because we were simply more comfortable using lam,
thus we'd be more glad if parallel programs had a version linked against
lam. :)

IMHO, it's always best to have as many alternatives as possible. That's
why free software is so widespread on Beowulf systems.


Eray (exa) Ozkural
Comp. Sci. Dept., Bilkent University, Ankara
e-mail: erayo@cs.bilkent.edu.tr
www: http://www.cs.bilkent.edu.tr/~erayo

Reply to: