Access To TESLA

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi

I am a MSc Student in Financial Engineering in London, and am planning to do a dissertation related to CUDA and parallel computing.
I already have access to GeForce 8800 but was wondering if there is a way to access TESLA hardware to be able to report improvements that it offers in my dissertation.
Does nVidia or any other company or organization (and no my university doesn’t at the moment ) provide such sort of free access to TESLA hardware?

Jambodevv

2

if you just want to show your improvement in GT200 and your problem size is not so large ( > 800 MB), then

I will suggest game cards, GTX275, GTX280 or GTX285, which are cheaper than TeslaC1060, but have the same

architecutre, GT200.

3

I agree with LSChien - either get a GTX260 core 216 or GTX285. I got a GTX 260 core 216 and its performance matches Tesla C1060 exactly (the 260’s slightly fewer cores balances out its higher memory bandwith).

Another choice is to try ZetaExpress for $25/month. See my review. warning: the service is a work in progress and if you’re not handy at working out compile problems, prepare for a bit of pain.