Thursday 3 September 2015

The Race Is On,

to build the world's fastest computer,


one that will beat China’s Tianhe-2 (above), from it's top spot where it has been for the past 5 times in the world tables, how fast is fast? a Linpack benchmark performance of 33.86 petaflops, or quadrillions of floating-point calculations per second, that's how fast, it is also the word from the 45th edition of the twice-annual TOP500 list of the world’s most powerful supercomputers, roughly twice as fast in the petaflops league as its nearest rival, a Cray XK7 a system installed at the Department of Energy’s (DOE) Oak Ridge National Laboratory, Titan, as it is known, is the top system in the United States, it achieved 17.59 petaflop/s on the Linpack benchmark, but look out as the Obama administration unveiled a new national initiative promising to deliver an exascale supercomputer 30 times more powerful than China’s Tianhe-2 by 2025, an exascale computer like the one envisioned by the Obama administration that's capable of one quintillion, or 1,000,000,000,000,000,000, or one billion billion floating-point operations per second or keeping it simple one exaflop, but is this a race you want to get into? back in 2010 PCAST, the President's Council of Advisors on Science and Technology, gave this advice, from the report, 

“While it would be imprudent to allow ourselves to fall significantly behind our peers with respect to scientific performance benchmarks that have demonstrable practical significance, a single-minded focus on maintaining clear superiority in terms of FLOPs count is probably not in our national interest,” 

in other words the so called 'arms race' in computers could be costly and an obsession with a FLOPs count could divert resources away from making other discoveries, all very well and good, but will it make writing our blog any quicker?


No comments: