Thomas Sterling and Satoshi Matsuoka on the Gap Between the TOP10 and Everything Else

Satoshi Matsuoka and Thomas Sterling

Over Primeur Magazine, Ad Emmen has posted a fascinating interview with Satoshi Matsuoka and Thomas Sterling on the topic of the growing gap between the Top10 systems and the rest of HPC.

The diversity is occurring now, but it is by no means the first time it has ever happened,” said Thomas Sterling. “As we look back over the last 40 years we see these different cases where the majority of people are working on one level with one modality and others are working at the very high end of the supercomputing. This time is different because this time both sides are still high-performance computing. Both are still pushing the edge beyond conventional enterprise servers or lap tops or desk side machines. One of the differences is that the two are out of phase with each other. The second issue is there are two very different objective functions. One is achieving sometimes competitively high performance, the other is accessibility. In case of accessibility, cost is a key issue.”

Read the Full Story.

Comments

  1. Michael Wolfe says

    Thomas Sterling asks whether there will be a trickle-down effect from HPC to the desktop (e.g., Intel MIC in PCs). HPC mostly benefits by the trickle-up effect from desktops: the economy of scale of desktop and workstation computing lowers the cost of commodity parts to a point where they are overwhelmingly cost-competitive with any custom HPC parts. Satoshi Matsuoka raises the question of whether this trickle-up will continue to be sustainable, as commodity parts vendors transition to focus on the mobile industry where performance is much less important than cost and power. As chip vendors change their investments to focus on the mobile growth market, who is going to pay the $100M or higher cost for the high performance processors needed in HPC? Will we just have to make do with a billion Quark or ARM cores?