[Beowulf] cluster softwares supporting parallel CFD computing

[Greg Lindahl]

On Tue, Sep 19, 2006 at 01:33:34PM +0200, Toon Knapen wrote:

> >So my real applications that show message rate is important don't
> >count for some reason?
> 
> Do you have some papers that describe in what way the overall 
> performance of the application depends on differences in message rate etc?

No, but I can show you a series of slides with performance compared to
Mellanox, and a histogram of the actual message sizes and rates of the
programs. In the case where the messages are short and frequent, we
end up with a large performance advantage. So if that's what your app
does when it scales, you can draw your own conclusion.

-- greg