(Excerpted from: “The Laboratory in the News,” Science &
Technology Review, Lawrence Livermore National Laboratory,
January/February 2000.)
On October 28, 1999, Lawrence Livermore and IBM celebrated the
“coming of age” of their Blue Pacific supercomputer with a special
ceremony at the Laboratory. This machine, part of the Department of
Energy’s Accelerated Strategic Computing Initiative (ASCI) has been
developed and delivered in several stages over the past three
years. It has become a mature and powerful tool for maintaining the
safety, reliability, and performance of the nation’s nuclear
stockpile.
Created by IBM, Blue Pacific performs at nearly 4 trillion
operations per second, applying all of its 5,856 processors in
parallel to a single computational problem. The massive
supercomputer is connected by nearly 8 kilometers (5 miles) of
cable and occupies an area covering some 750 square meters (8,000
square feet).
With all of the critical elements—software and code development, a
functional problem-solving environment, interconnect and
communication capabilities, data storage facilities—now in place,
ASCI’s Blue Pacific has emerged as a fully functional supercomputer
to help fulfill the requirements of DOE’s Stockpile Stewardship
Program.
The October ceremony included presentations by Laboratory Director
Bruce Tarter, DOE Deputy Assistant Secretary for Research,
Development, and Simulation Gilbert G. Weigand, and IBM Senior Vice
President for Technology and Manufacturing Nicholas M. Donofrio. A
sampling of breakthrough research calculations performed on the new
computer was highlighted.
The ceremony also included a preview of Option White, currently
being built by IBM as an extension of Blue Pacific. Able to perform
10 trillion operations per second, Option White has three times the
capacity and capabilities of Blue Pacific. It is scheduled for
demonstration in March 2000, with delivery to Lawrence Livermore
planned for the summer of 2000.
