LAM/MPI logo

LAM/MPI General User's Mailing List Archives

  |   Home   |   Download   |   Documentation   |   FAQ   |   all just in this list

From: Gustavo Seabra (gustavo.seabra_at_[hidden])
Date: 2008-11-13 12:40:54


On Thu, Nov 13, 2008 at 8:52 AM, Brian Barrett wrote:
> Hi everyone -
>
> This is probably a dangerous thing to do, it has been 15 months since our
> last LAM release and 5 months since the last change in the beta, so I'm
> considering doing a LAM release around the end of December, if people are
> interested. I'd need some beta testers to really put it through its paces,
> so please let me know if there is interest in the release.
>
> The only thing that's on the known issues I'm going to fix list is some
> changes to environment variables used by the SLURM batch scheduler. So if
> there are issues you'd like to see fixed, please let me know. If you have
> patches, that would be even better. I can't promise I'll fix any reported
> bugs, but we'll see where we end up.
>
> Of course, if I hear silence, I'll assume everyone's using Open MPI and not
> do the release ;).
>
> Thanks,
>
> Brian
>

One issue that I consider important is: Is there going to be support
for it after the release? I ask that because after a new version is
released, if people start using it and running into problems with it,
they will expect some sort of help... "If they released a new version
then there must be someone working on it, etc..."

It did happen to me quite recently to have problems trying to install
LAM/MPI in Cygwin, which is listed as a supported platform. As I run
into trouble, I tried to get help and noticed that it was quite hard,
specially since most developers had already moved on to OpenMPI. I
finally went ahead and moved to OpenMPI myself. (Still have trouble,
but at least there I have someone to talk to...)

So, unless there's some plan to keep supporting it, I'd suggest *not*
to bring a new release, except for occasional bugfixes which will
benefit whoever is already using it.

There it is... You can deposit my 2c in my acount ;-)

Cheers,

-- 
Gustavo Seabra
Postdoctoral Associate
Quantum Theory Project - University of Florida
Gainesville - Florida - USA