X / CUDA on CentOS, multi-gpu servers

[Bruce Labitt]

List is quiet...

Not quite sure how to ask this question.  Please forgive my inexpert wording and description.

We have run into a small issue getting X running on our multi-gpu server with CUDA.  Running CenOS 5.9.
Reading the forum on devtalk.nvidia.com yields some clues.   They involve blacklisting nouveau in grub, and some other things.

Has anyone set up a system like this before?   Can you help us get our system going?

What would it take to get X and CUDA to play nice?  How much time?  Anyone available to do this?  We want to remote in to the server and have the display directed to our local machines.  I have an X server on my local windows machine, and of course I have it on my local linux machine.

I don't think it is that hard, (especially since I've done it with an ordinary server) but I'd rather a professional (that is one of you, I hope) do this for us.

You can contact me off-list for any quotes.  Or, let me know on list that this is so simple, and this is how one does it.  I have found many bits and pieces on the web, but this needs to work soon, because this is preventing $WORK$ from happening.


Thanks,
Bruce

Bruce Labitt
Radar Systems Engineer
Autoliv Active Safety
1001 Pawtucket Blvd
Lowell, MA 01853
Tel: 1-978-674-6526
bruce.labitt at autoliv.com


***************************************************************
Consider the environment before printing this message.

To read Autoliv's Information and Confidentiality Notice, follow this link:
http://www.autoliv.com/disclaimer.html
***************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.gnhlug.org/mailman/private/gnhlug-discuss/attachments/20131014/98754af1/attachment.html