[Paraview] Sanity Check - Parallel GPU rendering

[Paul McIntosh]

> I think the Overview cat of the VirtualGL's documentation will 
> me more comprehensible than my english : 
> http://www.virtualgl.org/vgldoc/2_2evolving/#hd002 
> 
> So this will able you to use as many as you want GPU for one 
> rendering, because mpi use only CPU, then I found a my test 
> machine (core i7 920 with a Quad-SLI of GTX295) that paraview 
> got faster rendering if he's launched with VirtualGL (rendering 
> by GPU) than standalone (rendering by CPU). Maybe my vision is 
> not exact, but in fact, performances are very different ! Let's 
> give a try to VirtualGL Shree :)

Thanks for you help Shree/Philippe

I am also working with VirtualGL (though I only have it working in the basic
configuration). Eventually the intention is to allow remote viz.

In the above are you talking about ParaView 3.8 RC with GPU acceleration
used i.e. comparing VirtualGL GPU rendering to ParaView GPU rendering? Or as
it reads with ParaView rendering in CPU (no GPU acceleration).

I know that pvserver uses the GPU because it fails to render if the correct
driver isn't present. Our system has all NVIDIA nodes - the ParaView 3.8 RC
GPU acceleration is far far quicker than software rendering. My impression
is that ParaView in parallel with GPU acceleration would be the fastest
solution.

Cheers,

Paul
---
www.internetscooter.com