Mar 01

Print this Post

Dresden named a CUDA Center of Excellence

The TU Dresden has been named a CUDA Center of Excellence by NVIDIA, the world leader in visual and high-performance computing.

Press release published on March 1, 2012

CUDA is NVIDIA’s parallel computing architecture that enables dramatic increases in computing performance by harnessing the power of the graphics processing units (GPUs). The NVIDIA CUDA Center of Excellence program recognises, rewards and fosters collaboration with leading institutions at the forefront of parallel computing research.

TU Dresden with its partners Helmholtz-Zentrum Dresden-Rossendorf (HZDR) and Max Planck Institute of Molecular Cell Biology and Genetics (MPI-CBG), acting under the common roof of the DRESDEN-concept e.V., joins 17 elite institutions worldwide that have demonstrated a unique vision for improving the technology and application of parallel computing, and empowering academics and scientists to conduct world-changing research. CUDA Center of Excellence members are provided with NVIDIA GPU equipment and research grants.

Members from the Center for Information Service and HPC (ZIH), TU Dresden and HZDR are leveraging NVIDIA GPUs to accelerate highly complex simulations of the inner workings of plasma, as they are used in modern cancer therapy and fusion reactors. As a result, researchers were able reduce time to process scientific calculations from weeks to a few days. In addition, by accelerating expensive and complex experiments in Biology and Radiology, system utilization is increased, enabling research in other areas to benefit from GPU computing. Due to these rich benefits enabled by GPU computing, Simunova, a developer of generic programming methods for computational sciences, such as the matrix template library (MTL), decided to join the center.

“GPU computing is an integral technology that enables the acceleration of time-critical simulations in the computational sciences”, says Prof. Dr. Wolfgang E. Nagel, TU Dresden – director of the center – and also adds: “I am impressed by the creativity of the scientists of the different research areas in using this modern technology. They help position Dresden as a leader in the ongoing footrace for new scientific advances.”

As part of this cooperation the scientists from the areas of biology, computer science, physics, mechanical engineering, medicine, and mathematics as well as the industry partners use synergies in research and development also in other research fields. Hence, they offer practical education in GPU accelerated computational simulations to students of TU Dresden. Furthermore, leading experts in various fields that also use GPU computing will present their work in monthly seminars and an annual workshop in Dresden to also offer a stage for discussion and further cooperation.

“NVIDIA congratulates TU Dresden and its partners on achieving this special recognition,” says Stefan Kraemer, NVIDIA sales director HPC – education. “TU Dresden is the first German institution to receive CUDA Center of Excellence status, joining other world-renowned universities and institutions that are using GPUs to accelerate scientific research, including the Barcelona Supercomputing Center, Cambridge University, Stanford University, and the University of Moscow.”

Research groups utilizing GPU computing at TU Dresden include: Prof. Cowan (HZDR, Radiation Physics), Dr. Gerbeth (HZDR, Fluid Dynamics), Dr. Gottschling (SimuNova), Prof. Gumhold (TU Dresden, Computer Science, Computer Graphics and Visualization), Prof. Hochberger (TU Dresden, Computer Science, Embedded Systems), Prof. Koch (TU Dresden, Medical Faculty), Prof. Lehner (TU Dresden, Computer Science, Databases), Prof. Nagel (TU Dresden, ZIH / Computer Science, Computer Architecture), Dr. Posselt (HZDR, Ion Beam Physics and Materials Research), Prof. Schroer (TU Dresden , Structural Physics), PD Dr. Stiller (TU Dresden, Fluid Dynamics), Prof. Tetzlaff (Tu Dresden, Fundamentals of Electrical Engineering), Prof. Voigt (TU Dresden, Computation Science), and Prof. Zerial (MPI-CBG).

The NVIDIA CUDA Center of Excellence program is competitive and prestigious, and any institution with a demonstrated commitment to both parallel computing research and education may apply for CCOE status. For more information visit: http://research.nvidia.com/content/cuda-centers-excellence.

Permanent link to this article: https://gcoe-dresden.de/dresden-named-a-cuda-center-of-excellence/