October 10, 2011
JHU institute named as CUDA Center of Excellence by NVIDIA
Johns Hopkins University has been named a CUDA Center of Excellence by NVIDIA, a California-based visual and parallel computing technology company, honoring the university’s pioneering use of GPU computing and the CUDA programming model across research within multiple science and engineering departments.
The Center of Excellence will be headquartered in Johns Hopkins’ Institute for Data Intensive Engineering and Science, bringing together the expertise of scholars from the Krieger School of Arts and Sciences, the Whiting School of Engineering and the Sheridan Libraries to develop tools and methods capable of mining knowledge from the colossal data sets being produced today. Scientists from the Space Telescope Science Institute, located on the university’s Homewood campus, are also partnering in the activities of the center.
“Modern scientific computing is developing an amazing diversity, where scientists assemble novel systems from combining commodity components in unusual ways,” said Alexander Szalay, the Alumni Centennial Professor in the Henry A. Rowland Department of Physics and Astronomy at Johns Hopkins. “Our collaboration with NVIDIA will open up new directions in data-intensive scientific computing, where we can stream petabytes of data at speeds of half a terabyte per second into an array of 100 GPU cards for processing at supercomputer rates, yet doing it at a cost affordable to a university lab.”
As a CUDA Center of Excellence, Johns Hopkins will utilize GPU computing equipment and grants provided by NVIDIA to support a number of research and academic programs, including deployment of the Data-Scope, a GPU-powered, ultra-high throughput supercomputer created to dramatically increase the speed of scientific analysis. A cross between a powerful microscope and a telescope for data, the Data-Scope will allow scientists from disciplines ranging from medicine and genetics to astrophysics, environmental science, oceanography and bioinformatics to examine and analyze enormous amounts of data from both “little picture” and “big picture” perspectives.
The CUDA Center of Excellence Program identifies, recognizes, rewards and cultivates collaboration with leading institutions at the forefront of parallel computing research. The program is highly competitive and prestigious, and the list of participants includes Harvard University, Stanford University, Georgia Tech, University of Maryland, University of Tennessee and others.
Along with the donation of a substantial amount of high-end computing equipment and the sharing of technology, the award includes a grant to underwrite an annual workshop. In addition, senior researchers from NVIDIA will visit Johns Hopkins for several days to collaborate and attend lectures.