Building super-computers seems a lot like drag racing: no matter how fast you are there’s always another mark to strive for. The Sandia and Oak Ridge National Labs just formed the Institutes for Advanced Architectures (of computers, not buildings). Their goal is to figure out how to build an exascale computer.
“An exascale computer is essential to perform more accurate simulations that,
in turn, support solutions for emerging science and engineering challenges in
national defense, energy assurance, advanced materials, climate, and medicine,”
said James Peery, director of the Computation, Computers, Information, and
Mathematics Center at Sandia.
As the name exascale implies, the systems will operate in the exaflop range,
where an exaflop is a quintillion—10 to the 18th power—operations per second. To
put this number in perspective, an exaflop system would deliver one thousand
times more operations per second than a petaflop system, which was the objective
of IBM’s Blue Gene project when it was
launched about eight years ago. (A petaflop is a quadrillion—10 to the 15th
power—floating point operations a second.)
Besides the Wow!-factor, so what? What’s that have to do with cancer? Well, the quote above mentions medicine as one of the areas that demands ever faster computers. Indeed, the field of biology is rapidly becoming one of the most data-intensive and, hence, compute-intensive fields there is. For instance, at the lecture Craig Venter gave the other night he showed a map of large numbers of previously unidentified genes he found by sampling sea water all over the world and sequencing the genes of millions of single-cell organisms that nobody has identified before. He said it took a million computer hours to crunch the data, and that’s just scratching the surface. Nobody knows, but it is possible that those genes may express substances that might eventually become relevant in cancer therapy. Continuing to explore the genetics of our full biological environment will require massive computing power; so will developing useful detailed computer models of how "healthy" cells and tissues work and comparing them to cells on their way to malignancy.