CISS is the working name of the project and it stands for Canadian Internetworked Scientific Supercomputer. It is pronounced the same way as "kiss".

Over the past four years, investments by the the Canada Foundation for Innovation has resulted in over $160 million in high-performance computing equipment being installed in Canada. These resources have had a huge impact in research productivity in this country. However, the equipment is distributed across the country at over 20 sites. The result is numerous "Canada-class" facilities, but no "world-class" resources. The top Canadian academic site is number 309 on the listing of the top 500 sites in the world.

Unfortunately, there are research problems that require access to computational resources that exceed that of any single site in Canada. Rather than look to other countries (notably the U.S.) for these cycles, CISS is an attempt to harness the research computing capacity in Canada to create a virtual supercomputer.

CISS-1 is giving Dr. Wolfgang Jaeger, one of this year's winners of a prestigious NSERC E.W.R. Steacie Fellowship, the chance to get a world-class research result in a single day. His application requires 3-6 years of computing time. Normally he would not tackle this problem since other researchers with access to more computing power would beat him the the result. And, of course, there is no prize for second place. By giving Dr. Jaeger's application 3-6 years of computer time in a single day, he can get get a world-class research result for Canada, that normally would not be possible.

The first CISS application is embarrassingly parallel -- filling in data points in a 3D grid. This application was deliberately chosen for its simplicity, since it was felt that the logistics of getting 1000s of processors working together the first time would consume a lot of time. It did -- and even then we underestimated how time-consuming it would be! We hope that CISS-1 is a prototype for future large-scale computations in Canada. The Trellis software's capabilities to handle more tightly-coupled forms of parallelism will be (hopefully) tested in a future national test of CISS (CISS-2).

If CISS-1 is a short-term vision, then we hope that CISS will become a long-term vision. Canada Foundation for Innovation requires that the computing sites share 20% of their resources. One can envision CISS being a monthly event where, for example, 3 days a month are set aside for large-scale national computations. This would be unique in the world, and a tremendous opportunity for Canadian scientists.

CISS is an attempt to build the software and social infrastructure for a Canada-wide metacomputer. We do not have any current plans to go beyond Canada.

The (eventually, open source) software for CISS is developed as part of the academic Trellis Project in the Department of Computing Science, University of Alberta.