Me too. We have been running Direct Numerical and Large-Eddy
Simulation of fluid turbulence on a 32 nodes (2 CPUs per node) Linux
Clusters. In addition, I used to use MPI in IBM SP system (up to 32
nodes). Only minor changes were added to my F90 code when changing from
IBM to Linux Clusters (about 1 day of programming work load). However, I
have no idea in parallel computing in Solaris and Windows 2000/xp.
From my point of view, mechanical engineering people not only
interested in parallel/mpi computing, but also the underneath numerical
methods (e.g. FEM, FDM or FVM) and the application of supercomputing in
engineering problems (e.g. fluid mechanics or solid mechanics). I would
be interested in large scale computational solution.
Hope it helps,
Chun-Ho LIU
Department of Mechanical Engineering
The University of Hong Kong
Feyzi Inanc wrote:
> We have a 64 node cluster running on Linux and our choice of MPI
> implementation is MPICH.
>
> At 02:24 PM 2/18/2003 +0000, you wrote:
>
>> I've been asked for a course on parallel
>> programming by our mech eng department.
>>
>> i was thinking of mpi and would look at
>>
>> solaris
>> linux
>> windows 2000/xp
>>
>> what implmentations are people on the list using at the moment?
>>
>> thanks in advance
>>
>> --
>>
>> Ian
>>
>> [log in to unmask]
>>
>> Home page
>>
>> http://www.kcl.ac.uk/kis/support/cit//fortran/
>>
>> comp-fortran-90 home page
>>
>> http://www.jiscmail.ac.uk/lists/comp-fortran-90.html
>
>
> Feyzi INANC, PhD
> ----------------------------------------------------------------------------------------------------
>
> Iowa State University, Center for Nondestructive Evaluation
> Applied Sciences Complex II, 1915 Scholl Road, Ames IA 50011
> ----------------------------------------------------------------------------------------------------
>
> Ph. (515) 294 9738 Fax. (515) 294 7771
> E-Mail. [log in to unmask]
> http://www.cnde.iastate.edu
> ----------------------------------------------------------------------------------------------------
>
>
|