ACADEMIA

CAPS entreprise, one of the leader in HPC compiler technology, today announce its first release supporting OpenACC. HMPP Workbench 3.1 is now supporting OpenACC version 1.0 standard. It shows CAPS commitment to the promotion of this standard and its willingness to make Many-Core programming easier.

OpenACC is an initiative from CAPS, CRAY, NVIDIA and PGI to provide the HPC community with a new open parallel programming standard. Based on a common set of directives for C and Fortran languages, OpenACC enables programmers to easily take advantage of the processing power of heterogeneous many-core architectures. Its multi-platform and multi-vendor compatible model offers a way to preserve investment in legacy applications by enabling an easy migration path to accelerated computing.

“The GPU computing breakthrough has allowed many users to propose new massively parallel codes to advance many scientific fields. With OpenACC we are simplifying the use of accelerators and leveraging legacy applications. We are very confident that this will help to further broaden the community taking advantage of many-core technologies.” said François BODIN, CAPS CTO.

“The goal of OpenACC is to provide a familiar directive-based programming model that makes it quick and simple for developers to take advantage of the performance benefits of accelerators like GPUs,” said Duncan Poole, president of OpenACC. “I congratulate CAPS on their latest release of HMPP Workbench, and applaud their efforts in helping us bring the tremendous value of directives and OpenACC to customers.”

More information about HMPP workbench is available at http://www.caps-entreprise.com