Score another one for IBM's supercomputing aspirations. The Department of Energy's National Nuclear Security Administration (NNSA) on Tuesday announced a contract with IBM to bring supercomputing systems to its Lawrence Livermore National Laboratory.
With the two new systems, IBM technology will play a role in continuing to ensure the safety and reliability of the nation's aging nuclear deterrent.
"The longstanding partnership of NNSA, Lawrence Livermore National Laboratory, and IBM is ushering in an era of multi-petaFLOPS computing," said NNSA Administrator Thomas D'Agostino. "These powerful machines will provide NNSA with the capabilities needed to resolve time-urgent and complex scientific problems, ensuring the viability of the nation's nuclear deterrent into the future. This endeavor will also help maintain U.S. leadership in high-performance computing and promote scientific discovery."
A 20-PetaFLOPS Supercomputer
IBM will deliver two systems. The first, called Dawn, a 500-teraFLOPS (trillion floating operations per second) BlueGene/P system, is scheduled for delivery in the first quarter. Dawn will lay the applications foundation for multi-petaFLOPS computing on Sequoia, the second system.
Sequoia, a 20-petaFLOPS (quadrillion floating operations per second) system based on future BlueGene technology, will be delivered starting in 2011 and deployed in 2012.
With a speed of 20 petaFLOPS, Sequoia is expected to be the most powerful supercomputer in the world and will be approximately more than 10 times faster than today's most powerful system. IBM puts it into perspective this way: If each of the 6.7 billion people on Earth had a hand calculator and worked together on a calculation 24 hours per day, 365 days a year, it would take 320 years to do what Sequoia will do in one hour.
Strengthening Predictive Simulation
Sequoia will have 1.6 petabytes of memory, 96 racks, 98,304 compute nodes, and 1.6 million cores. Though orders of magnitude more powerful than such predecessor systems as ASC Purple and BlueGene/L, Sequoia will be more than 160 times more power-efficient than Purple and 17 times more than BG/L.
Sequoia and Dawn will serve NNSA's tri-lab Advanced Simulation and Computing (ASC) program, which unites the scientific computing resources and expertise of Los Alamos, Sandia and Lawrence Livermore national laboratories.
The Sequoia system will focus on strengthening the foundations of predictive simulation through running very large suites of complex simulations called uncertainty quantification (UQ) studies.
The machines also will be used for weapons-science calculations necessary to build more accurate physical models. The NNSA said this work is a cornerstone of its Stockpile Stewardship program to ensure the safety, security and reliability of the U.S. nuclear-weapons stockpile today and into the future without underground testing.
Environmentally Friendly Nuclear Testing
Sequoia represents a strong example of the exponential leaps forward in supercomputing, according to Charles King, principal analyst at Pund-IT. With the type of architecture that BlueGene leverages, King expects to see this order of magnitude continue in coming years.
"These types of systems are usually being operated to run simulations of various and sundry types of nuclear events," King said. "Basically, scientists are doing in the brains of a supercomputer what they used to have to do by building a bomb, digging a hole in the desert and setting it off."
Not only will Sequoia complete the simulation faster and with greater levels of sophistication than past supercomputers, King said, it will also lessen the environmental impact of weapons testing without sacrificing the assurance that nuclear devices being designed or stored are safe and capable of doing the job they were designed to do.
"When it comes to the types of beneficial uses supercomputers can accomplish, the surface is only beginning to be scratched," King said. "But Sequoia is going to be scratching it pretty deep."