Communications of the ACM
Stanford Researchers Break Million-Core Supercomputer Barrier
A floor view of the newly installed Sequoia supercomputer at the Lawrence Livermore National Laboratories.
Credit: Lawrence Livermore National Laboratories
Researchers at Stanford University's Center for Turbulence Research (CTR) used a supercomputer with more than 1 million computing cores to solve a complex fluid dynamics problem. The researchers, using the Sequoia IBM BlueGene/Q system at Lawrence Livermore National Laboratories, say their work proves that million-core fluid dynamics simulations are possible. The simulation examined predicting the noise generated by a supersonic jet engine, and the research aims to design quieter aircraft engines.
"Only recently, with the advent of massive supercomputers boasting hundreds of thousands of computing cores, have engineers been able to model jet engines and the noise they produce with accuracy and speed," says CTR director Parviz Moin.
Computational fluid dynamics simulations test all aspects of a supercomputer. Supercomputers such as Sequoia divide the simulation's complex math into smaller parts so they can be computed simultaneously. The more cores that are available, the faster and more complex the calculations can be. "These runs represent at least an order-of-magnitude increase in computational power over the largest simulations performed at the Center for Turbulence Research previously," says CTR researcher Joseph Nichols. "The implications for predictive science are mind-boggling."
From Stanford University
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found