Hi Derek,

On Tue, Jun 24, 2008 at 1:30 PM, Derek Gaston <[EMAIL PROTECTED]> wrote:
>
> Here is how much each proc is using:
>
> #CPU : #MB/proc
> 256 : 200-700
> 128 : 350-700
> 64 : 450-800

Thanks!  These numbers are pretty interesting.  I know there are some
places in the code where we go ahead and build a serial std::vector of
size global n_elem or n_nodes.  I think this happens sometimes on all
processors and sometimes on just one, so it's possible we will run out
of memory on one CPU way before the others...

>
> First thing to note is that the #MB/proc is a _range_.  This range is
> taken from me watching "top" on one of the compute nodes.  The memory
> usage of each process _oscillates_ between the two numbers listed
> about every 5 seconds.  Based on watching using "xosview" I believe
> that the high numbers occur during communication steps, while the low
> numbers are during "assembly" (residual computation).  At this point,
> these are just guesses, but I've been watching these things for a few
> weeks now and kind of have a feel for what's going on.

I believe mpip (http://mpip.sourceforge.net/) can report more detailed
memory usage, but implementing the profiling does require additional
work.  I'd be interested to see how uniform the memory usage is across
all the procs.  We have some people around here who know about mpip so
if you find some strange things happening I might be able to take a
closer look.

Also, it sounds like now might be a good time to start thinking about
our StructuredMesh implementation again...unless your eventual problem
will be unstructured and you don't care.  We could probably run some
truly enormous matrix-free problems on structured grids....

-- 
John

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Libmesh-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-devel

Reply via email to