| layout | title | permalink |
|---|---|---|
page |
About |
/about/ |
In the last ten years, there has been an increased drive toward the use of many-core architectures which exploit larger amounts of data parallelism. GPUs have evolved from a fixed pipeline graphics processing hardware into powerful programmable co-processing units capable of performing general purpose computing (sometimes referred to as GPGPU or General Purpose Computing on GPUs). In comparison with traditional CPU systems, a GPU is capable of far higher (theoretical) peak performance within a smaller power window and as such many of the Top 500 supercomputers are reliant on GPU (or other many-core) architectures.
There is an active research of community within the University of Sheffield developing novel techniques for GPUs. There is also a growing community of researchers using GPUs for Deep Learning. Despite this there is a clear skills gap between those who are able to utilise these highly parallel architectures and those that can not. The GPUComputing Sheffield site aims to act as a University of Sheffield community portal to promote skills and software for GPU and accelerated computing. More specially the webiste give University of Sheffield members access to permanent training material, information and advice about using and buying hardware and information of software packages for using and developing programs for GPUs.
The GPUComputing@Sheffield group has been an NVIDIA GPU research centre since 2011 and an NVIDIA Education Centre since 2016 (more history). The group is maintained mainly through through a mailing-list and this website which provides information on all aspects of GPU programming, software, hardware and consultancy.