IBM announced it has build a new "hybrid" supercomputer with AMD Opteron and Cell engines for the Department of Energy's National Nuclear Security Administration. This new supercomputer is called Roadrunner and it's world's first system that
can operate at one petaflop (one thousand trillion calculations per second). That's twice as much as the current number one supercomputer, the IBM Blue Gene system at Lawrence Livermore National Lab.
IBM says Roadrunner will be used at the Los Alamos National Laboratory to ensure the safety and reliability of the nation's nuclear weapons stockpile. Additionally, the system will also be used for astronomy, energy, human genome science and climate change research.
Roadrunner cost about $100 million and is world's first hybrid supercomputer, it features 6,948 dual-core AMD Opteron chips as well as 12,960 Cell engines.
Made from Commercial Parts. In total, Roadrunner connects 6,948 dual-core AMD Opteron chips (on IBM Model LS21 blade servers) as well as 12,960 Cell engines (on IBM Model QS22 blade servers). The Roadrunner system has 80 terabytes of memory, and is housed in 288 refrigerator-sized, IBM BladeCenter racks occupying 6,000 square feet. Its 10,000 connections – both Infiniband and Gigabit Ethernet -- require 57 miles of fiber optic cable. Roadrunner weighs 500,000 lbs. Companies that contributed components and technology include; Emcore, Flextronics, Mellanox and Voltaire.
Compared to most traditional supercomputer designs, Roadrunner’s hybrid format sips power (3.9 megawatts) and delivers world-leading efficiency – 376 million calculations per watt. IBM expects Roadrunner to place among the top energy-efficient systems later in June when the official “Green 500” list of supercomputers is issued.
IBM is developing new software to make Cell-powered hybrid computing broadly accessible. Roadrunner’s massive software effort targets commercial applications for hybrid supercomputing. With corporate and academic partners, IBM is developing an open-source ecosystem that will bring hybrid supercomputing to financial services, energy exploration and medical imaging industries among others.