NVIDIA has two major markets these days: gamers and scientists. While the company unveiled its latest GPUs last month to the delight of gamers, today at the ISC Conference for high-performance computing in Germany the company unveiled a beta version of the new compiler for the OpenPOWER platform.
The OpenPOWER Foundation, of which NVIDIA is a member, will foster a new compiler for OpenACC applications capable of running on POWER processors. The compiler is currently in beta and can be downloaded by requesting a pass from the PGI website.
(Related: NVIDIA invests in AI)
This compiler is a product of the Portland Group, otherwise known as PGI, which NVIDIA purchased in 2013. PGI claimed its first beta version will support PGI Fortran, C and C++ on Linux-based OpenPOWER machines.
The new OpenACC compiler will allow developers to build applications that take full advantage of GPUs (and now POWER CPUs) to construct high-performance computing applications. The compiler will be available to anyone in August, but until that time, potential users must request access.
NVIDIA also announced successful hackathons for OpenACC, as well as a few new members for the OpenACC support organization. Chief among them was the University of Illinois at Urbana-Champaign.
John L. Larson, research scientist at the University of Illinois, said, “UIUC joined OpenACC to support two key grants and projects from the NSF/NCSA and [Department of Education] that involve scientific research using complex, un-optimized code. Being a member of the OpenACC standard organization will give both projects early access to technical features of the developing OpenACC language and also enable input that may influence new OpenACC language features.”
