NVIDIA launches new research, training and certification programmes for developers focused on GPU computing

1 Jun 2010 Hamburg - The ecosystem surrounding the NVIDIA CUDA architecture for parallel processing got a significant boost with the establishment of new programmes to advance the field of general-purpose computing on graphics processing units (GPGPU). With thousands of research papers already published, and more than 350 universities teaching CUDA, a figure that has tripled over the past year, these new programmes will expand the teaching and use of GPUs.

The new programmes include:

  • CUDA Certification Programme: Driven by the demand for qualified GPGPU engineers, this is the first programme to certify expertise in massively parallel programming on GPUs.
  • CUDA Research Centers: This programme recognizes institutions that embrace GPU Computing across multiple research fields.
  • CUDA Teaching Centers: This formally recognizes institutions that have integrated GPU Computing techniques into their mainstream computer programming curriculum.

In addition, NVIDIA is launching an all new NVResearch online portal, providing information on global research projects supported by NVIDIA, as well as details on all the education and research oriented programs run by the group.

"There are now more than 350 universities worldwide teaching the CUDA programming model within their curriculum, and more than 100,000 programmers actively developing applications that use the GPU", stated Sanford Russell, general manager of CUDA & GPU Computing at NVIDIA. "These new programmes will encourage this work and develop collaborations that will advance GPGPU adoption across a wide variety of industries."

The CUDA Certification programme responds to the industry's demand for qualified parallel programmers. To become an NVIDIA CUDA Certified Engineer, candidates must demonstrate good working knowledge of the CUDA architecture and programming model, ability to apply CUDA constructs to common algorithmic frameworks and strong understanding of optimization techniques to get the most performance from CUDA C based code.

The CUDA Research Center programme recognizes and fosters collaboration with research groups at universities and research institutes that are expanding the frontier of massively parallel computing. Among the benefits are exclusive events with key researchers and academics, a designated NVIDIA technical liaison and access to specialized online and in-person training sessions.

The CUDA Teaching Center programme is the first of its kind to be developed and offered to universities and colleges by a hardware vendor. The programme has many benefits, including the donation of teaching kits consisting of text books, software licenses and CUDA-enabled GPUs for teaching lab computers as well as academic discounts for additional hardware if required.

"CSIRO was one of the first supercomputing centers to combine GPUs and CPUs and run applications up to 200 times faster as a result", stated Dr. John Taylor, leader of the Computational and Simulation Sciences platform at CSIRO. "We are honoured to be named a CUDA Research Center as we look forward to continuing to leverage our CSIRO GPU cluster. It will enable CSIRO, in a cost effective way, to be globally competitive in addressing computational challenges for 'big science'."

The CUDA Research Center and CUDA Teaching Center programmes will have global reach with institutions across Asia, Europe and North America. Already selected as CUDA Research Centers are John Hopkins University (U.S.), Nanyan University (Singapore), Technical University of Ostrava (Czech Republic), CSIRO (Australia) and SINTEF (Norway). Already selected as CUDA Teaching Centers are McMaster University (Canada), State University of New York, Potsdam (U.S.), California Polytechnic State University, San Luis Obispo (U.S.), ITESM (Mexico), Czech Technical University (Czech Republic) and Qingdao University (China).

These new programmes augment the existing CUDA Center of Excellence program, the elite network of ten prominent institutes focused on advancing parallel computing on the GPU. They are Cambridge University, Chinese Academy of Sciences, Harvard University, University of Maryland, National Taiwan University, Tokyo Tech, Tsinghua University, University of Illinois at Urbana-Champaign, University of Tennessee and University of Utah.
Source: NVIDIA