Two cards on the same cluster


I’m newbie
I am working on a small project
There are two Nvidia GPU’s connected to the cluster

When run the program the program is executed on default slow GPU

I want it to be executed on the faster one

How do we select the Faster GPU to be enabled.

Is it to mentioned somewhre in the NVidia sdk files

the two dev ices are

GeForce 8600 GTS"

Please let me know
Thanks in advance

You can use “cudaSetDevice()” to choose the device to run on. See the programming guide for details.

Thank you very much Simon
that was really helpful :)