San Diego Supercomputer Center, Intel Collaborate on Earthquake Research
New software, 10+ teraflops to better predict earthquake paths in southern CaliforniaResearchers from the San Diego Supercomputer Center (SDSC) and Intel have developed new software that has resulted in the fastest seismic simulations ever, which will better predict ground motions during earthquakes.
The collaboration, which helps researchers better predict paths of earthquakes in southern California, is aimed at saving lives and minimizing property damage, according to a recent SDSC press release.
The team of researchers ran seismic simulations at 10.4 petaflops per second by using 612,000 Intel Xeon Phi processor cores on the new Cori Phase II supercomputer at the National Energy Research Scientific Computing Center. That beat the previous record of 8.6 petaflops per second for seismic simulations, which were run on China’s Tianhe-2 supercomputer, the release said.
In earthquake research, it’s critical to obtain higher frequencies to predict ground motions. And through the efficient use of the latest and largest supercomputers, such as Cori, seismologists are able to “increase the frequency content of the simulated seismic wave field,” the release said.
“In addition to using the entire Cori Phase II supercomputer, our research also showed a substantial gain in efficiency in using the new software,” said Alex Breuer, a postdoctoral researcher from SDSC’s High Performance Geocomputing Laboratory.
SDSC and Intel researchers developed the new seismic software package, called EDGE (Extreme-Scale Discontinuous Galerkin Environment), during the past year as part of a collaboration that began with the creation of SDSC’s Intel Parallel Computing Center.
“Research and results such as the massive seismic simulation demonstrated by the SDSC/Intel team are tremendous for their contributions to science and society,” said Joe Curley, senior director of Intel’s Code Modernization Organization. “Equally, this work also demonstrates the benefit to society of developing modern applications to exploit power-efficient and highly parallel CPU technology.