[Paraview] GPU acceleration on Paraview

[Kiyani, Khurom]

Hello,

I am very much a novice when using paraview. I am doing some plasma turbulence simulations and have been trying to volume render my data cube (about 5GB size). On linux it took ages, slowed down my graphics, and then finally crashed my system but on windows the rendering happened very fast (5 minutes) but I could not visualize properly (zoom in/zoom out) afterwards -- paraview stalled incredibly. I couldn't do anything after this. I have a pretty good workstation with 2 six core Xeon processors (24 threads in total), with 192 GB ram and two NVIDIA cards (a Tesla K20c with 5GB ram and around 2500 cores, and a NVIDIA Quadro 2000 card for the visualisation).  Could you please tell me what I can do to improve the rending time and solve the problem of post-rendering visualisation. In particular I think it was only using 4 of the cpu-cores. I have heard on many forums that GPU hardware acceleration can significantly speed things up -- especially with my Tesla. I have the Quadro and Tesla cards running in a NVIDIA maximus configuration with the Quadro on dedicated graphics and the Tesla on dedicated compute.

Your help is much appreciated.

With many thanks and kind regards,

Khurom

*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*
"Be Excellent to Each Other"
- Bill & Ted -
Khurom Kiyani
Laboratoire de Physique des Plasmas
Ecole Polytechnique
Route de Saclay
91128 Palaiseau
Paris, France
Tel: +33 (0)7 81 94 27 73
URL: http://go.warwick.ac.uk/kkiyani
*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20130927/2c928583/attachment-0001.htm>