For Immediate Release
Innovative Computing Technique Facilitates Unprecedented Simulation and Earns Livermore Team the Gordon Bell Prize
Reno, NV — November 15,2007 — Using groundbreakingcomputational techniques, a team of scientists from Lawrence Livermore NationalLaboratory and IBM earned the 2007 Gordon Bell Prize with a first-of-a-kindsimulation of Kelvin-Helmholtz instability in molten metals on BlueGene/L, theworld’s fastest supercomputer.
Sponsored by ACMand the IEEE Computer Society, SC07 showcases the latest advances highperformance computing, networking, storage and analysis. SCinet serves as theplatform for SC07 exhibitors to demonstrate advanced computing resources fromtheir home institutions and elsewhere by supporting supercomputing and gridcomputing applications.
By performing extremelylarge-scale molecular dynamics simulations, the team was able to study, for thefirst time, how a Kelvin-Helmholtz instability develops from atomic scalefluctuations into micron-scale vortices.
“This has never been donebefore. We were able to observe this atom by atom. There was no time scale orlength scale we couldn’t see,” said Jim Glosli, lead author on the winningentry titled “Extending Stability Beyond CPU Millennium: A Micron-ScaleSimulation of Kelvin-Helmholtz Instability.” Other team members were: KyleCaspersen, David Richards, Robert Rudd and project leader Fred Streitz of LLNL;and John Gunnels of IBM.
The Kelvin-Helmholtzinstability arises at the interface of fluids in shear flow and results in theformation of waves and vortices. Waves formed by Kelvin-Helmholtz (KH)instability are found in all manner of natural phenomena, such as waves on awindblown ocean, sand dunes and swirling cloud billows. While Kelvin-Helmholtz instability hasbeen thoroughly studied for years and its behavior is well understood at themacro-scale, scientists did not clearly understand how it evolves at the atomicscale until now. The insights gained through simulation of this phenomenon areof interest to the National Nuclear Security Administration’s (NNSA) StockpileStewardship Program, the effort to ensure the safety security and reliabilityof the nation’s nuclear deterrent without nuclear testing. Understanding howmatter transitions from a continuous medium at macroscopic length scales to adiscrete atomistic medium at the nanoscale has important implications for suchLaboratory research efforts as National Ignition Facility (NIF) laser fusionexperiments and developing applications for nanotube technology.
“This was an importantsimulation for exploring the atomic origins of hydrodynamic phenomena, andhydrodynamics is at the heart of what we do at the Laboratory,” Glosli said. “Wewere trying to answer the question: how does the atomic scale feed into thehydrodynamic scale.”
"This remarkableKelvin-Helmholtz simulation breaks new ground in physics and in high performancescientific computing," said Dona Crawford, associate director forComputation at Lawrence Livermore National Laboratory. "A hallmark of theAdvanced Simulation and Computing program is delivering cutting edge sciencefor national security and the computing that makes it possible."
This simulation ofunprecedented resolution was made possible by the innovative computationaltechnique used –a techniquethat could change the way high performance scientific computing is conducted.Traditionally the hardware errors or failures that are an inevitable part ofHPC have been handled by the hardware itself or the operating system. Thisstrategy was perfectly adequate for 1,000 to 10,000 processor supercomputingsystems. However, these traditional approaches don’t work as well on amassively parallel machine the size of BG/L with over 200,000 CPUs (centralprocessing units) -- almost 10 times more than on any other system. With such alarge number of processors and components, hardware failures are almost certainduring long production runs. Hardware failures impact system performance andconsume valuable time on the machine.
In partnership with IBM, theLivermore team pioneered a new strategy for recovering from hardware failure.They developed a way to use the application itself to help correct errors andfailures. Their reasoning was that the application, which has a completeunderstanding of the calculation being run, can evaluate the errors and decidethe most efficient strategy for recovery. For example, by implementing a strategy to mitigate cache memory faults(which are the primary cause of failure in BG/L), the team was able to runwithout error for CPU-millennia.
“Applications with thiscapability could potentially lead to a new paradigm in supercomputer design,”said Streitz, noting that application-assisted failure recovery reduceshardware reliability constraints, opening the way for supercomputer designsusing less stable but higher performing – and perhaps less expensive –components. “That concept may allow the building of a faster machine.”
Named for one of thefounders of supercomputing, the prestigious Gordon Bell Prize is awarded toinnovators who advance high-performance computing. The award is widely regardedas the Oscars of supercomputing. A Livermore team led by Streitz won the 2005Gordon Bell Prize for a simulation investigating the solidification in tantalumand uranium at extreme temperatures and pressure, with simulations ranging insize from 64,000 atoms to 524 million atoms. This year on the expanded machine,the Livermore team was able to conduct simulations of up to 62.5 billion atoms.
“The scale of thisKelvin-Helmholtz simulation was enormous compared to the previous simulations,”Streitz said. “We were really pushing the limits of what is currently possibleon this machine.”
SC07, sponsoredby ACM and IEEE Computer Society, will showcase how high-performance computing,networking, storage and analysis lead to advances in research, education andcommerce. This premiere international conference includes technical andeducation programs, workshops, tutorials, an exhibit area, demonstrations andhands-on learning. For moreinformation, please visit http://sc07.supercomputing.org/.