It may come as a surprise to many who are deeply engaged in exascale research
and development that is has already been six years since the first U.S. community
town hall meetings were convened to discuss the challenges for the next level of
computing. It was in Spring 2007 when meetings were held at Berkeley, Argonne
and Oak Ridge that formed the basis for the first comprehensive look at exascale.
Even more surprising is that, in spite of numerous national and international
activities, the community has not made any significant progress towards reaching
the goal of an exaflops system. If one reflects on early projections, it seemed
possible to build at least a prototype by 2020. I believe that the lack of progress
in the intervening years has made it all but impossible to see a working exaflops
system by the end of this decade. Specifically, I don't not expect a working
exaflops system to appear in the #1 spot of the TOP500 list with an RMAX performance exceeding 1 exaflop/s by November 2019. In this talk, I will explain why this is a regrettable lack of progress and define the major barriers.