The University of Sheffield named an NVIDIA GPU Education Center
The University of Sheffield announced that it has been named a GPU Education Center by NVIDIA, a computer technology company that has pioneered GPU-accelerated computing.
GPU Education Centers are recognised institutions that have integrated GPU-accelerated computing into their mainstream science and engineering curriculum. GPU-accelerated computing leverages the parallel processing capabilities of GPU accelerators and enabling software to deliver dramatic increases in performance for scientific, artificial intelligence, deep learning, graphics, engineering, and other demanding applications.
The University of Sheffield was recognised for its commitment to advancing the state of parallel programming education.
More specifically, Dr Paul Richmond (Computer Science), a recently appointed EPSRC Research Software Engineering (RSE) Fellow, has been leading Sheffield’s initiative to embed GPU-accelerated computing into mainstream education and research. The recently created COM4521 (and Masters COM6521) module on “Parallel Computing with GPUs” provides a highly desired training of how to write efficient parallel programs and how to utilize GPU architectures to accelerate large-scale scientific programs. The module will be delivered on an annual basis to undergraduates and is also available to staff and postgraduate research students.
Dr Richmond is also instigating short courses and training on the use of GPUs for scientific research. The training consists of one or two day interactive sessions which make GPU computing available for a broad range research domains. These courses will be delivered at Sheffield, but will be advertised nationally and provided to non-University of Sheffield members without charge. Current partners supporting GPU education at Sheffield include EPCC, Oxford e-Science, the Science Technology and Facilities Council, the Software Sustainability Institute and the N8 consortium. Each partner will help promote the schedule of nationally available GPU training events hosted at Sheffield. This will include the development of cloud-based teaching resources, video recorded lectures and material to help other research software engineers teach GPU computing at their own academic institutions. University of Sheffield members should consult the GPUComputing@Sheffield and external members can receive information via the Archer training portal.
Dr Richmond says: “GPU-accelerated computing plays a vital role in improving the scale and performance of simulation within research. It is vitally important that our students gain these essential insights so that they can go out into industry and academia to create software that is able to scale to increasing levels of parallelism. I am delighted that the University of Sheffield has been named an NVIDIA Education Centre, as my own plans for increasing the level of GPU education are increased and it is wonderful to have NVIDIA’s support.”
“The University of Sheffield, and Dr Richmond in particular, have long been advocates of the use of GPU-accelerated computing in research,” said Kimberly Powell, director of Higher Education at NVIDIA. “The GPU Education Centre designation reflects the significant commitment and investment Sheffield has made in this area, and we look forward to working with Dr Richmond to prepare new waves of students to harness the power of GPU-accelerated computing to drive innovation and scientific discovery.”
As a GPU Education Center, The University of Sheffield will have access to NVIDIA GPU hardware and software, and NVIDIA parallel programming experts and resources, including educational webinars and an array of teaching materials. The University of Sheffield students and faculty also receive access to free GPU programming training at nvidia.qwiklab.com. The comprehensive, self-paced training spans introductory to expert level coursework on a range of GPU programming languages, applications, libraries, programming models and tools. GPU Education Centers were formerly known as CUDA Teaching Centers.