Nvidia is a founding member of Stanford University’s new Pervasive Parallelism Lab (PPL). The PPL will develop new techniques, tools, and training materials to allow software engineers to harness the parallelism of the multiple processors that are already available in virtually every new computer.
Nvidia’s investment complements the company’s ongoing strategy to solve some of the world’s most computationally intensive problems with its GPUs, tools and software. The company has enjoyed significant success to date with its Tesla line of GPU computing hardware solutions and, more importantly, with CUDA technology, its programming environment that gives developers access to the massively parallel architecture of the GPU through the industry-standard C language.
‘Parallel programming is perhaps the largest problem in computer science today and is the major obstacle to the continued scaling of computing performance that has fueled the computing industry, and several related industries, for the last 40 years,’ said Bill Dally, chair of the computer science department at Stanford.
Until recently, computer installations delivering massive parallelism could only be deployed in large-scale computer centres with hundreds to thousands of separate computer systems. With the recent introduction of many-core processors such as the GPU and the multi-core CPU, most new computer systems come equipped with multiple processors that require new software techniques to exploit parallelism. Without new software techniques, computer scientists are concerned that rapid increases in the speed of computing could stall.
From fundamental hardware to new user-friendly programming languages that will allow developers to exploit parallelism automatically, the PPL will allow programmers to implement their algorithms in accessible, domain-specific languages while at deeper, more fundamental levels of software, the system would do all the work for them in optimising the code for parallel processing.
Nvidia GPU technology, combined with the CUDA programming environment, has delivered speed increases anywhere from 8× to 50× over conventional processing technologies. Nvidia joins with AMD, Hewlett Packard, IBM, Intel, and Sun Microsystems in this venture.
