POWERB16
27-11-2007, 02:28:15
System NameBlueGene/L SiteDOE/NNSA/LLNL (http://www.top500.org/site/2556)System FamilyIBM BlueGeneSystem ModeleServer Blue Gene SolutionComputereServer Blue Gene SolutionVendorIBMURLhttp://www.llnl.gov/asc/comput... (http://www.llnl.gov/asc/computing_resources/bluegenel/bluegene_home.html)Application areaResearchMain Memory32768 GBInstallation Year2005 Operating SystemCNK/SLES 9Memory32768 GBInterconnectProprietary (http://www.top500.org/connfam/18)ProcessorPowerPC 440 700 MHz (2.8 GFlops)
http://www.top500.org/files/systems/bluegene_photo.jpghttp://www.top500.org/files/systems/bluegene_photo.jpg
(http://www.top500.org/system/7747#thumb)
Housed in Lawrence Livermore National Laboratory’s Terascale Simulation Facility, BlueGene/L (BGL) clocked 478.2 trillion floating operations per second (teraFLOPS) on LINPACK, the industry standard of measure for high-performance computing. Built by IBM, BGL is a workhorse supercomputer used to make possible science simulation of unprecedented detail for NNSA’s tri-lab Advanced Simulation and Computing (ASC) Program, which leverages the computing expertise and resources of Sandia, Los Alamos and Lawrence Livermore national laboratories. Computer simulations are a cornerstone of NNSA’s program to ensure the safety, security and reliability of the nation’s nuclear deterrent without underground testing – stockpile stewardship.
Recently expanded to accommodate growing demand for high-performance systems able to run the most complex nuclear weapons science calculations, BGL now has a peak speed of 596 teraFLOPS. In partnership with IBM, the machine was scaled up from 65,536 to 106,496 nodes in five rows of racks; the 40,960 new nodes have double the memory of those installed in the original machine.
The upgrading of BGL, notably through the addition of nodes with twice the memory, allows scientists from the three nuclear weapons labs to develop and explore a broader set of applications than the single package weapons science oriented work that has been the mainstay of the machine in the past. For example, BGL had been used widely for materials science calculations such as assessing materials at extreme temperatures and pressures. Now it will be much easier to run more complex applications related to modeling integrated systems as opposed to focused exploration of one area of physics or chemistry.
http://www.top500.org/files/systems/bluegene_photo.jpghttp://www.top500.org/files/systems/bluegene_photo.jpg
(http://www.top500.org/system/7747#thumb)
Housed in Lawrence Livermore National Laboratory’s Terascale Simulation Facility, BlueGene/L (BGL) clocked 478.2 trillion floating operations per second (teraFLOPS) on LINPACK, the industry standard of measure for high-performance computing. Built by IBM, BGL is a workhorse supercomputer used to make possible science simulation of unprecedented detail for NNSA’s tri-lab Advanced Simulation and Computing (ASC) Program, which leverages the computing expertise and resources of Sandia, Los Alamos and Lawrence Livermore national laboratories. Computer simulations are a cornerstone of NNSA’s program to ensure the safety, security and reliability of the nation’s nuclear deterrent without underground testing – stockpile stewardship.
Recently expanded to accommodate growing demand for high-performance systems able to run the most complex nuclear weapons science calculations, BGL now has a peak speed of 596 teraFLOPS. In partnership with IBM, the machine was scaled up from 65,536 to 106,496 nodes in five rows of racks; the 40,960 new nodes have double the memory of those installed in the original machine.
The upgrading of BGL, notably through the addition of nodes with twice the memory, allows scientists from the three nuclear weapons labs to develop and explore a broader set of applications than the single package weapons science oriented work that has been the mainstay of the machine in the past. For example, BGL had been used widely for materials science calculations such as assessing materials at extreme temperatures and pressures. Now it will be much easier to run more complex applications related to modeling integrated systems as opposed to focused exploration of one area of physics or chemistry.