This is an old revision of the document!
GPU Configuration
If you reboot teslas, be sure to run the initialization script:
sudo /opt/nvidia/config/cuda_init_script.sh restart
If you reboot fermis, be sure to run the initialization script:
sudo nvidia-smi -a
CUDA & MPI on GPUs
Most recent cuda & sdk installation: /opt/nvidia/cuda4: cuda4.0 toolkit for teslas and fermis /opt/nvidia/cuda4/NVIDIA_GPU_Computing_SDK: cuda sdk 4.0 for teslas and fermis
legacy cuda & sdk installation: /opt/nvidia/cuda: cuda2.3 toolkit for teslas and fermis /opt/nvidia/cuda-2.3_sdk: cuda sdk 2.3 for teslas and fermis
The working MPI is in /opt/MPI/mpich2-1.2.1-install/ All the includes/libs/executables are in there, and to use it with fortran, make sure you have the fortran compiler /opt/intel/Compiler/11.1/059/bin/intel64/bin/ifort
