Blue waters - The furure of Xtreme Scientific Computation

[Randy]

I'm forwarding an announcement from the eScience mailing list. I'll be there!
-----------------------------------------------------------------------------------------------------------
Thom Dunning is the director of NCSA (home of the IBM Blue Waters
petascale machine), and one of the world's leading computational
chemists.

Thom will be at UW on Tuesday February 22 and Wednesday 23.

He'll be giving a talk in Guggenheim at 4 on the 22nd.  I've attached
an announcement, and it's on the eScience blog.
-----------------------------------------------------------------------------------------------------------

[Randy]

When I posted the notice for this talk, I inadvertently called it the "furure" of Xtreme computing rather than the "future" of Xtreme computing. As it turn out, neither title is appropriate. If I understood the talk correctly, Blue Waters will represent the end of the road, rather than the future.

Dr. Dunning's talk outlined the hardware used to create Blue Waters. Here is a link describing the hardware specs: Blue Waters Hardware.

Basically, Dr. Dunning said that Blue Waters will reach the limits of what is possible with current approaches, due to energy consumption, cost, and reliability. He also mentioned that the current software for high-performance computing is not always up to the task. For example, compilers that are unable to take advantage of the architecture of a machine like Blue Waters.

Blue Waters will be used for all of the applications you might expect, which makes it a computer suited to a broad range of data and computation intensive applications, but I'm wondering if that makes it a jack of  all trades, master of none? In terms of the problems we would like to solve in biology, I'm curious as to why we aren't designing a machine that is purpose built for protein folding, for example.

As an aside, there was a small joke made at the end of the talk that seemed to disparage the idea of quantum computing. While I don't pretend to understand the possibilities or likelihood of quantum computing, I do think that the future of Xtreme computing lies in some novel approach. For a quick look at what is being done in the UK, check out Susan Stepney's research home page

I do thank the UW's eScience department for sponsoring the talk. They seem to be a progressive and open department that is addressing cutting edge computational problems.