Intending to increase the efficiency of writing parallel code, NVIDIA, Cray, The Portland Group and CAPS enterprise recently announced OpenACC, a new open parallel programming standard targeting millions of scientific and technical programmers wanting to tap into the power of CPU/GPU computing systems.
Programming professionals performing research in industries as diverse as chemistry, biology, physics, analytics, weather and intelligence stand to benefit from this new paradigm which allows them to take advantage of “directives” – simple hints which expose parallelism to the compiler, instead of having to modify existing code, as the compiler will take care of mapping the computation onto the accelerator.
The added beauty of this approach lies in the fact that the code base is compatible across platforms from multiple vendors, taking advantage of the time and money spent developing legacy programs. How powerful is this new programming model? Developers are reporting performance gains ranging from 2X to 10X in as little as two weeks.
For example, a global manufacturer of navigation systems wanted to speed up its research involving algorithms for real-time object detection in image sequences. Using a combination of directives and NVIDIA GPUs, they were able to accelerate their production application by 5x in just one week.
A postdoctoral researcher at the University of Texas at San Antonio is on a mission to better understand the effects of solvents on proteins and accelerated his application 5x in a single day.
“People are sometimes afraid to try GPUs, but once they try them out, they realize it’s much easier than they initially expected.” – NVIDIA’s Sumit Gupta
“OpenACC represents a major development for the scientific community,” said Jeffrey Vetter, joint professor in the Computational Science and Engineering School of the College of Computing at Georgia institute of Technology. “Programming models for open science by definition need to be flexible, open and portable across multiple platforms; OpenACC is well designed to fill this need. It provides a valuable new tool to empower the vast numbers of domain scientists who could benefit from application acceleration, but who may not have the funding or expertise to port their code to emerging architectures.”
Assisting Military Applications
As you may recall, NVIDIA was selected by the Defense Advanced Research Projects Agency (DARPA) to participate in the Ubiquitous High Performance Computing (UHPC) program and prototype systems are due by 2018. In the coming years a combination of UHPC and OpenACC could serve to accelerate the operational capability of our intelligence agencies and all branches of the Armed Forces.
The UHPC program aims to develop computer systems, from embedded to cabinet level, which have extremely high energy efficiency and are dependable and easily programmable. These systems will have dramatically reduced power consumption while delivering a thousand-fold increase in processing capabilities.
With a wide range of mission critical government and defense applications such as pattern recognition, video processing and stabilization, cryptography, sensor fusion, target tracking and flight simulation in need of the significant performance enhancements provided by GPU Computing, Trenton continues to develop integrated rackmount computers based on our strategic partnership with NVIDIA.
