Accelerating the adaptation of scientific simulation codes to increasingly powerful supercomputers
LLNL's Todd Gamblin in front of the Sequoia supercomputer

Lawrence Livermore Computation Researcher Awarded Early Career Funding

Wednesday, May 14, 2014

LLNL scientist Todd Gamblin, a computer scientist in the Center for Applied Scientific Computing, has been selected by the U.S. Department of Energy’s Office of Science Early Career Research Program to receive funding for a proposed project entitled “Statistical Methods for Exascale Performance Modeling.”

Todd will receive up to $2.5 million in funding over five years for a project to accelerate the adaptation of scientific simulation codes to increasingly powerful supercomputers, a process that currently can take up to six months for complex applications. Increasingly complex machine architectures and applications are making this process even slower. His research is particularly important as the high-performance computing (HPC) community prepares to ramp up computing speeds from petascale (quadrillions of floating point operation per second) to exascale systems that will be as much as 1,000 times faster. HPC experts believe the first exascale systems will come online in the 2020 timeframe.

The dynamically changing behavior of modern simulation codes makes existing techniques for modeling their performance difficult to apply. Todd proposes to develop statistical models of applications that can represent adaptive, data-dependent code behavior in a manner that can be scaled up for more powerful computing systems. In addition, the project will develop techniques to reduce the complexity of application models, so that application developers understand them. “The idea is for these new models to provide simulation developers with insights that allow them to quickly optimize the performance of their code, ensuring that applications can take full advantage of the performance of future exascale machines,” he said.

Todd earned his M.S. and Ph.D. in computer science from the University of North Carolina in Chapel Hill. He first came to Lawrence Livermore as a student in 2008, returned as a post-doc in 2009 and was hired as staff in 2010. For his undergraduate work, Gamblin completed a double major in computer science and Japanese at Williams College in Massachusetts. Subsequently, he worked as a software developer in Japan and held graduate research internships at the University of Tokyo and IBM Research.

“Disbelief” was how he initially reacted to learning he had been selected for an award by DOE’s Advanced Scientific Computing Research program. “It seemed like slim odds,” Gamblin said. “I’m grateful and I’m happy to be able to work on projects I’ve had on the back burner as part of the early career program.”

Learn more about Todd Gamblin.