[aspect-devel] Error in moving temporary files during output

Timo Heister heister at clemson.edu
Mon Jan 13 10:34:34 PST 2014


> I am
> currently doing scaling tests on a Cray XC30 up to 4000 cores and
> getting the same error messages.

by the way: can you share your setup (compiler versions and flags you
give to cmake, etc.)?

> I am writing in a designated $WORK
> directory that is intended for data output, however there is also a fast
> local $TMPDIR directory on each compute node.
> I guess my question is: Is the Number of grouped files = 1 parameter
> always useful in case of a large parallel computation and an existent
> MPI I/O system or is this cluster specific (in that case I will just
> contact the system administrators for help)?

So, "Number of grouped files = 1" will use MPI I/O, and "Number of
grouped files = 0" will just write separate files (standard posix)
into tmp and then move them over. It depends on your filesystem what
MPI I/O does. It might be that you gain nothing (if it is a standard
NFS filesystem), or it will be MUCH faster (we are talking several
gigabyte/s on a parallel filesystem).

> Another thing I would like to mention is that this system only allows
> for static linking.

I have no experience with that.

> Could somebody who introduced the possibility
> to dynamically load libraries at runtime comment on the work to make
> this a switch at compile-time?

Dynamically loading libraries won't work if you link statically, but
we could introduce a cmake flag to disable this code.

-- 
Timo Heister
http://www.math.clemson.edu/~heister/


More information about the Aspect-devel mailing list