Parallel FDS

[Kevin McGrattan <kevin.mcgrattan@nist.gov>]

To those interested in parallel processing:

Thanks for the feedback on your experiences in compiling
and/or running the beta
parallel version of FDS 4. From the wide variety of
responses it is clear that we
need to do some more work documenting, explaining,
debugging. The good news is that
some of you were able to run, and for those that didn't, the
error statements helped
us clean things up a bit.

Could those of you who tried to run in parallel and who 
haven't already contacted me about it, report to me the
following

Did you try to compile the source code? If yes, what
compiler/OS?

Could you successfully run a job? If no, what type of error
are you getting?

A few things to keep in mind -- the number of processors you
request must be
exactly the same as the number of meshes in your
calculation. If you are using
only one grid, you can use only one CPU to compute.

The working directory has to be shared or accessible by all
the machines.


Thanks

Kevin



-- 
Kevin McGrattan
National Institute of Standards and Technology
100 Bureau Drive, Stop 8663
Gaithersburg MD 20899-8663

Telephone: (301) 975 2712
FAX:       (301) 975 4052