Which is the most powerful computer in the world? Easy: the Tianhe-1A at Tianjin in China. Just ask the chaps at the TOP500 supercomputer list – or President Barack Obama.
"Just recently, China became home of ... the world’s fastest computer," said the prez in his latest State of the Union speech.
That makes the high performance computing profs at the University of Florida pretty mad. They say their machine Novo-G would actually, in some circumstances, kick the Tianhe-1A's arse.
“Novo-G is believed to be the most powerful reconfigurable machine on the planet and, for some applications, it is the most powerful computer of any kind on the planet,” insists Professor Alan George, director of the National Science Foundation’s Center for High-Performance Reconfigurable Computing (CHREC).
George says the TOP500 rankings don't mean a whole lot as they are based on the LINPACK benchmark, which he contends is irrelevant to many supercomputing problems today.
A top-10 supercomputer weighs hundreds of tonnes and uses megawatts of power, yet the Novo-G is "the size of two home refrigerators" – quite small even when one's talking about American fridges. And it consumes a measly 8 kilowatts. How can such a tiny machine rival the brobdingnagian behemoths at the head of the TOP500 list?
According to George and his colleagues, the answer in reconfigurable processors whose logic architecture can be optimised for an application rather than remaining fixed as in conventional machines. The Novo-G with its 192 reconfigurable processors "can rival the speed of the world’s largest supercomputers at a tiny fraction of their cost, size, power, and cooling”, according to a recent article by the prof and his colleagues.
We learn from a Florida uni statement issued yesterday that Novo-G is to get another 192 special flexi-processors later this year, a process which will apparently involve only a small increase in power consumption, cooling etc.
Presumably if it now "rivals" the world's largest supercomputers, following the upgrade the Novo-G will leave them behind conclusively – at least in its favoured applications such as "genome research, cancer diagnosis, plant science, and the ability to analyze large data sets".
Meanwhile those willing to subscribe can read all about the radical new reconfigurable future here courtesy of IEEE Computing in Science and Engineering magazine
Thursday, February 24, 2011
Subscribe to:
Posts (Atom)