New York University Faculty of Arts and Science College of Arts and Science Graduate School of Arts and Science

OpenCL/AMD_APP GPU Computing Servers

All the CIMS Linux machines are Intel OpenCL compatible.

Test it out by running:

cp -r /usr/local/pkg/opencl/tools-master ./
cd tools-master
make
./print-devices
./cl-demo 1000000 10

In addition, CIMS has two servers running that are are powering 4 AMD Fire GL V7800s that are allocated for OpenCL/AMD APP (Accelerated Parallel Processing) SDK usage.  Each system is connected to 2 GPUs.  These systems are capable of both single and double floating point precession calculations.

The hardware is available on opencl1.cims.nyu.edu and opencl2.cims.nyu.edu.  These machines are restricted to logins from within the Courant network, so if you are coming from outside of CIMS, you will have to first login to access.cims.nyu.edu, then use ssh to get to them.

Once logged in you can get setup running OpenCL/AMD APP code by following the instructions above.  You'll be able to choose between Intel and AMD OpenCL implementations.

See the AMD APP SDK doucmentation for  more information.

If you run into problems, please let us know as we'd like to try and get all the kinks worked out.