UCL has been named a GPU Research Centre by NVIDIA following a successful campus-wide application. Researchers across campus came together to put in a UCL-wide bid, with the application led by Dr Matt Clarkson (Translational Imaging Group, Centre for Medical Image Computing), Dr James Hetherington (UCL Research Software Development Group), Daghan Cam (The Bartlett School of Architecture) and Prof Sebastien Ourselin (UCL Institute of Biomedical Engineering).
UCL was recognised for its advanced research in a wide variety of areas, including bio-molecular modelling for cancer research, vascular imaging and magnetic resonance physics, computational systems biology, tsunami modelling, image registration, finite element modelling and image guided surgery, to name a few. In addition, as a member of the SES (Science and Engineering South) consortium, a world-leading science and engineering research hub formed by 5 leading research-intensive universities, UCL researchers have access to the EMERALD GPU cluster, one of the largest in Europe.
Daghan Cam, Leader of GPU Computing Research at The Bartlett School of Architecture, says: ‘This college wide initiative and partnership with NVIDIA is enabling cross-disciplinary collaborations and accelerating our research in many areas like real-time simulations, computational physics and deep learning, as well as practical applications such as computer vision for robotics and 3D printing.’
GPU-accelerated computing is the use of a GPU together with a CPU to accelerate scientific, analytics, engineering, consumer, and enterprise applications. Computational scientists have used the dramatic performance increases provided by accelerated computing to tackle the research computing challenges of the 21st century.
As a GPU Research Centre, UCL will have pre-release access to NVIDIA GPU hardware and software that will enable it to reach new levels of performance and continue delivering advanced innovative research across disciplines. GPU Research Centres were formerly known as CUDA Research Centres.
Read more
Daghan Cam is a Tutor in the MArch AD programme’s Wonderlab.
UCL Bartlett School of Architecture
Centre for Medical Image Computing
Image: MArch AD, Wonderlab, RC1, Cellular Division simulated via CUDA GPU-run supercomputing. Research directed by: Alisa Andrasek & Daghan Cam. Students: Danli Zhong, Jong Hee Lee, Ningzhu Wang, Feng Zhou.