The University of Toronto has announced that their new $50 million supercomputer has finally come to life. The ~300TFlop IBM machine is built using IBM System X iDataPlex servers. 30,240 cores [2.53Ghz] are said to chew nearly $1million per year in electrical energy. Cooling is handled by traditional methods in the summer and the Canadian winters during the colder months.
It’s a really impressive, world-class facility,” said Chris Loken, the chief technical officer for the SciNet consortium, which is responsible for getting the supercomputer built. “It’s going to give a lot of scientists access to some really powerful resources.”
It will not be in the top 20 systems next year,” Mr. Loken said. “But there’s still going to be a lot of awfully good research that can be done on it.
All in all, its an impressive machine. One of the more interesting facts about the system is its contiunous system management. The system management software will continuously manage the nodes and shut down any that have been idle for more than ten minutes. How this situation interacts with the traditional batch system is beyond the scope of this artcile, so I’m interested to learn more.
For more info, read the full article here.