There are fast computers, and then there’s IBM’s BlueGene/L
supercomputer.
The Department of Energy’s National Nuclear Security Administration
(NNSA) and IBM today announced the world’s fastest supercomputer has outdone
itself. In its latest testing, IBM said its BlueGene/L (BG/L) achieved a
sustained performance of 207.3 trillion floating-point operations per second
(teraFLOPS), a new record for floating-point performance.
IBM said its software researchers were able to improve
performance with new mathematical libraries that take better advantage of
the dual-core architecture of the floating point unit of the PowerPC 440
processor used in Blue-Gene/L. A year ago, BG/L’s floating point performance
was less than 100 teraFLOPS.
The BG/L had already been recognized by the Top500.org as the world’s fastest
supercomputer, reaching a speed of 367 teraFLOPS in overall performance. The
independent Top500.org is expected to release new rankings later this month.
The IBM powerhouse 64-rack system is driven by a whopping 131,072 PowerPC 440
processors.
BG/L is housed at NNSA’s Lawrence Livermore National Laboratory. The
record speed was achieved on Qbox computer code for conducting materials
science simulations critical to national security. Qbox is a first
principles molecular dynamics (FPMD) code, designed to predict the
properties of metals under extreme conditions of temperature and pressure –
a long-standing goal for researchers in materials science and high
energy-density physics. These FPMD codes are used for complex simulations
at the atomic level in a number of scientific areas, including metallurgy,
solid-state physics, chemistry, biology and nanotechnology.
The computer simulation capabilities provide the nuclear weapons analysis
that NNSA said it needs to keep the nuclear weapons stockpile safe, secure
and reliable without underground nuclear testing.
“This is an important step on the path to performing predictive
simulations of nuclear weapons, and these simulations are vital to ensuring
the safety and reliability of our nuclear weapons stockpile,” said Dimitri
Kusnezov, head of NNSA’s Advanced Simulation and Computing (ASC) program.
The capability to do predictive science is key to NNSA’s national
security mission, as its researchers try to understand how the materials in
nuclear weapons age, particularly for those warheads that have aged beyond
their intended life. IBM said the performance of the Qbox code, specially
designed to run on large-scale platforms such as BG/L, has implications for
the broader research community and will likely enable the development of new
materials of interest to many industries.
“Disruptive advanced architecture work for ASC leads to low-cost, but
highly useful computers that benefit the nation well beyond national
security,” said Kusnezov.
Future application areas might include super-enhanced simulations in
other areas.
“Take, for example, a pharmaceutical company designing a drug for the
heart,” said Jim Sexton, a PhD researcher at IBM’s TJ Watson Research Center
in Yorkstown, N.Y. “In a few years they might be able to model the heart and
how the drug interacts with it over a period of several weeks,” Sexton told
internetnews.com.
Similarly, Sexton sees the potential for car manufacturers to model a new
engine and simulate wear and tear over its expected lifetime and optimize
performance.
“BlueGene is showing the way,” said Sexton. “Now the commercial world is
asking questions they would never have considered a computer being able to
answer.”
