Computer Chess Club Archives


Search

Terms

Messages

Subject: Will Blue Gene be Obsolete before it is Completed?

Author: Timothy J. Frohlick

Date: 10:29:11 02/28/00

Go up one level in this thread


On February 28, 2000 at 12:37:54, Dan Simmons wrote:

>Excerpt from article in Physician's Financial News:
>
>IBM is following up on the success of its chess-playing supercomputer Deep Blue
>with a new endeavor that will delve into the mysteries of the human body.  The
>company recently said it plans to build a computer of staggering power to solve
>the mystery of the structure of proteins, the workhorse molecules and building
>blocks of the body.
>
>Paul Horn, senior vice-pres of reasearch at IBM, said the computer, called Blue
>Gene, could provide crucial understanding of how viruses like hepatitis and HIV
>attack the body....
>
>IBM expects it will take four to five years and $100 million to build Blue Gene,
>which will be a million times faster than the average desktop computer.  It will
>perform 1 quadrillion mathematical operations per second - 500 times more than
>the fastest computer today.
>
>Still, IBM's task is daunting....
>
>Blue Gene will have 1 million processors, the central computing engines of
>computers, working together.  The concept is not new, but the scale is
>unprecedented and will force the computer to be "self-healing" - that is, it has
>to be able to detect failing components, seal them off and direct the work
>elsewhere.
>
>In 1997, IBM's Deep Blue supercomputer defeated the world's greatest chess
>player, Garry Kasparov, in a highly publicized tournament.

Sounds Impressive,

Five years to do this project seems a bit much.  Will processor enhancements
render Blue Gene obsolete before it gets going?  This sounds like a NASA
project.  If each processor consumes two watts then we can all keep nice and
warm at least.


Tim Frohlick



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.