xengine performance on RS6000

[Jeff Smith]

In <1991Apr17.202630.28470 at eagle.lerc.nasa.gov> fsfrick at bones.lerc.nasa.gov (David Fricker) writes:

>In a related vein, we've done some 'benchmarking' using local programs
>on an RS6000 model 530 & various SGI Irises.  The relative performance
>figures are _very_ strange.  On one CFD program doing a 2D grid, an
>SGI Personal Iris (4D/25) outperforms my RS6000 model 530, while a 
>3D grid run has my model 530 outperforming a 4D/340VGX Iris! (The 340
>means the iris has 4 processors.)  Both programs were written by the
>same person & solve roughly the same problem--vortex sheet rollup--but
>one program solves it in 2D while the other solves it in 3D.

>Any ideas as to why these relative performance figures are so strange?

The 3d model probably is very floating point intensive, which is where
the RS/6000's really shine.  Does the application on the 4D/340VGX
take advantage of having multiple CPUS?  If not, it will have a
tough time keeping up to the 530.

jeffs at sgi.com