[mpich-discuss] (no subject)

[Mandar Gurav]

Hi Ehsan!

Since your codes are running correctly on i7 based system, you can
easily (as it is) run your codes on 4 Xeon processor based system.

If you are considering to run it on multiple number of nodes/systems
they you need to some additional settings. But, for single system with
4 Xeons - don't worry at all.

-- Mandar Gurav

On Tue, Nov 29, 2011 at 12:12 PM, ehsan saei <shajarian_e at yahoo.com> wrote:
> hi all
> I'm graduate student of physics so I am not a good programmer I wrote codes for numerical computations and because the speed is important for me I studied about parallel processing and find that mpi is a good library for what I want (my computations are not independent so I need communication at some points for example a[i]=a[i]+b[i-1] ) I studied about mpi library and wrote some parallel codes and I got correct answer from my codes   and saw that the runtime decreased by increasing the number of processes .I did this works on my pc with core i7 cpu but now I want to work on a single computer with 4 intel xeon processors (hp proliant dl380) my question is that can mpi use all performance of my system?
> _______________________________________________
> mpich-discuss mailing list     mpich-discuss at mcs.anl.gov
> To manage subscription options or unsubscribe:
> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss



-- 
Mandar Gurav,
Junior Research Fellow,
High Performance Computing Laboratory,
Department of Electrical Engineering,
Indian Institute of Technology Bombay,
Powai, Mumbai
Phone: +91-22-25764408 (Office)
http://www.mandargurav.org