The client application is comparatively lightweight, so I think  you can
generally expect better performance for data intensive tasks with N server
processes instead of N-1. Note, you can run more than N if you wish, but you
likely will see performance degrade sharply.

Of course how it scales in practice depends greatly on the problem. What
ParaView scales well at is data intensive problems, where it benefits most
because there is N times the available RAM when it is run on a cluster with
N nodes, an to a lesser degree on having N CPUs to run through the data.

David E DeMarle
Kitware, Inc.
R&D Engineer
28 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-371-3971 x109


On Thu, Feb 11, 2010 at 4:20 AM, bassaidai <[email protected]> wrote:

> Hi all,
> if we have a workstation with N processor (on wich we run paraview and
> pvserver in two separate shells) must we run pvserver in parallel on a
> number of processors strictly smaller than N?
> For example, if we have N=4 and in the first shell we run ParaView (with
> "paraview"), in the other shell can we use maximum 3 processors (with
> "mpirun -np 3 pvserver") or can we use all processors (with "mpirun -np 4
> pvserver")?
>
> Thank you very much
>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>
>
_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Reply via email to