The University of Guelph will take part in a nationwide supercomputer test Monday, Nov. 4 that will link thousands of computers together across Canada to solve an important computational chemistry question in one day -- a task that would normally take six years to complete.
The project is the first test of the Canadian Internetworked Scientific Supercomputer (CISS), an initiative that attempts to harness the research computing capacity in Canada to create a virtual supercomputer. Monday’s effort will involve researchers and facilities from 18 Canadian universities and will create the most powerful computer in the country for a day. The virtual supercomputer will work on research devised by University of Alberta chemist Wolfgang Jaeger to calculate energy shifts as two molecules are manipulated around 3-D space.
Computing and information science professor Deborah Stacey and SHARCNET system administrator John Morton will oversee Guelph’s role in the unique demonstration. The University will have 108 computer processors helping work on the problem. “This is the kind of research that requires access to computational resources that exceed that of any single site in Canada,” Stacey said. “We will have computers running in almost every province, all of them working on the same problem. But all of the computers will continue to be ‘working machines,’ which means that their normal daily functions will not be disrupted.”
U of G is taking part in Monday’s test through SHARCNET (Shared Hierarchical Academic Research Computing Network), a network of high-performance Beowulf computer clusters at seven Ontario post-secondary institutions: Guelph, the University of Western Ontario, McMaster, Wilfrid Laurier University, the University of Windsor, Fanshawe College and Sheridan College. SHARCNET represents 20 per cent of all the supercomputing power available in Canada and is the most powerful computer in the country available for use by university researchers.
Future CISS efforts include helping unravel the mysteries of the human genome, predicting long-term weather patterns and understanding how to build compounds and devices that are a billionth of a metre in size. More information about the CISS project is available on the website: http://www.cs.ualberta.ca/~ciss.
Cite This Page: