SANTA CLARA, Calif.--()--The OpenACC standards group today announced growing support for OpenACC-supported development tools, and initial results from programmers who have been using the recently-released OpenACC compilers to accelerate research.
“Compilers that support OpenACC directives play an important role in developing portable software”
OpenACC is a programming standard for parallel computing on accelerators using directives, designed to enable millions of scientists around the world to easily take advantage of the transformative power of heterogeneous CPU/GPU computing systems. It is now available in compiler products from the OpenACC founding members, Cray, The Portland Group (PGI) and CAPS enterprise. It is also gaining increasing support in other programming tools, including recently released solutions by Allinea and RogueWave, which provide visual debugging of OpenACC directives on Cray XK6 systems.
OpenACC provides the easiest way for scientists, with or without extensive parallel programming expertise, to accelerate their research in a matter of hours using familiar programming models.
Researchers at NASA Ames Research Center’s, NAS Division, which plays a critical role supporting America’s space and aeronautic programs, have been using OpenACC-supported compilers and are already experiencing a significant acceleration of their applications with limited effort:
“Using PGI’s OpenACC compiler, we ported a computational fluid dynamics (CFD) application benchmark to a general purpose GPU-based system,” reported NASA researchers in an upcoming research paper. “OpenACC is a much easier way to accelerate applications than other programming approaches, and we saw an immediate speed up of the benchmark on multiple tests, up to 10X faster compared with a single CPU core-based system.”
University of Houston Building OpenACC Test Suite
To help developers of OpenACC compilers, the University of Houston is launching an OpenACC feature test suite. Scheduled to be released in the summer of 2012, the suite will test expected functionality of OpenACC features and will help speed development of new, robust products. The test suite can be downloaded from the OpenACC website when available.
“Compilers that support OpenACC directives play an important role in developing portable software,” said Dr. Barbara Chapman, professor of Computer Science at the University of Houston. “By leveraging our experience with parallel programming models and tools, and OpenMP in particular, we were able to quickly identify what is required to support the OpenACC members.”
“OpenACC is quickly becoming the parallel programming model of choice for scientific researchers using hybrid computers, because directives are simple to use and transparent when compiled for non-accelerated platforms,” said Duncan Poole, president of the OpenACC. “OpenACC is anticipated to benefit a broad range of scientists working in chemistry, biology, physics, data analytics, weather and climate, intelligence, and many other fields.”
OpenACC Highlighted at NVIDIA GPU Technology Conference
The OpenACC programming standard will be highlighted at a number of sessions at the upcoming NVIDIA GPU Technology Conference (GTC 2012) – May 14-17, San Jose, Calif. – an annual event that aims to advance global awareness of GPU computing and its importance to the future of science and innovation. OpenACC will be featured in a number of presentations and hands-on tutorials designed to educate programmers and scientists on the programming model and best practices, including:
- Monday, May 14, 10:30am, 1pm, 2:30pm – Programming GPUs with OpenACC (Parts 1, 2 and 3; Presented by NVIDIA and PGI)
- Monday, May 14, 1pm, 2:30pm – Programming Heterogeneous Many-cores Using Directives (Parts 1,2; Presented by CAPS)
- Tuesday May 15, Tuesday, May 15, 4pm – Accelerator Directives, OpenACC and OpenMP4ACC (Presented by Cray)
- Wednesday, May 16, 3:00pm – How to Bake Portable Many-Core Programs (Presented by CAPS)
- Thursday, May 17, 9:30am – NVIDIA OpenACC
- Thursday, May 17, 10am – The PGI Fortran and C99 OpenACC Compilers
For a complete list of OpenACC sessions and room locations at GTC 2012, visit the GPU Technology Conference Scheduler web site and search on “OpenACC.”
For information about other upcoming OpenACC events, visit the events calendar on the OpenACC web site.
OpenACC Product Availability
Multiple OpenACC-supported compilers are now available from CAPS, Cray and PGI. PGI released a version of the PGI Accelerator™ compiler (v. 12.4) in April 2012, which provides select OpenACC support for scientists using Fortran or C programming languages. CAPS enterprise released its HMPP Workbench 3.1 solution with OpenACC support in April 2012. Cray delivered initial support for OpenACC in its [NAME] compiler, which was released in December 2011. More OpenACC capabilities and features are expected to be available from all three companies in future releases.
For more information about the new compilers from CAPS, Cray and PGI, visit the following websites:
About OpenACC API
The OpenACC Application Program Interface describes a collection of compiler directives to specify loops and regions of code in standard C and Fortran to be offloaded from a host CPU to an attached accelerator, providing portability across operating systems, host CPUs and accelerators. OpenACC allows programmers to provide simple hints (directives) to the compiler, identifying which areas of code to accelerate. By exposing parallelism to the compiler, directives allow the compiler to do the detailed work of mapping the computation onto the accelerator. OpenACC enables users to create a single code base that runs on heterogeneous many-core accelerators as well as multi-core systems, making scaling application performance easier and more portable than ever. It also offers an ideal way to preserve investment in legacy applications.
For more information about OpenACC visit the OpenACC.org web site.
OpenACC is a non-profit corporation founded by the group of companies that developed the OpenACC Application Program Interface specification. OpenACC was formed to help create and foster a cross platform API that would allow any scientist or programmer to easily accelerate their application on modern many-core and multi-core processors using directives. For more information, visit www.openacc.org.