Nvidia’s Sumit Gupta writes that the recent flap over the $Billion-dollar cost of the Japanese K Supercomputer brings to light that supercomputers do not have to cost nearly this much to develop and operate.
The 112 billion Yen price tag for the K computer was just the start. Add to this about 10 billion Yen ($128 million U.S.) each year to power and maintain the mammoth system, and it’s clear that the costs will really start to add up. The K system costs so much to build because the SPARC CPU at the heart of the machine is an expensive, custom-designed processor. HPC history has repeatedly shown that the development cost of custom processors is just not economically viable in the high-performance computing market.
Gupta contrasts the K supercomputer with Tsubame 2.0, a GPU-powered machine at the Tokyo Institute of Technology. He calls it “a great example of a high-performance, cost-efficient system” that currently ranks as the #5 system on the Top500 list.
Read the Full Story.