Dear Discovery Cluster Users:

IF YOU DO NOT USE GPU QUEUES ON THE DISCOVERY CLUSTER PLEASE IGNORE THIS EMAIL MESSAGE:

There have been many requests for OpenACC 2.0 accelerated capability on our GPU nodes. In response to this I now have GCC 5.2.1 compiled with OpenACC 2.0 and NVIDIA CUDA SDK 7.0. It is now possible to compile and run OpenACC/CUDA/OpenMP/MPI code on our GPU nodes.

Documentation on compiling and using OpenACC with CUDA is here - http://nuweb12.neu.edu/rc/?page_id=18#openacc .

The PDF document (in the link above) can also be downloaded from here - http://nuweb12.neu.edu/rc/wp-content/uploads/2015/09/GNU-5.2_OpenACC-2.0_cuda-7.0_discovery_gpu-nodes.pdf .

Thanks,
Best
Nilay

########################################################################

To unsubscribe from the DISCOVERY list, click the following link:
http://listserv.neu.edu/cgi-bin/wa?SUBED1=DISCOVERY&A=1