Can supercomputers help societies prepare for earthquakes?

At least some Intel Researchers will tell you “Of course!”! Their confidence received a strong boost this week when their work with Technische Universität München and Ludwig-Maximilians-Universität München: Petascale High Order Dynamic Rupture Earthquake Simulations on Heterogeneous Supercomputers earned the honor of being one of the five finalists for the most coveted award in the world of supercomputing, ACM Gordon Bell Prize. This prize acknowledges an outstanding achievement in applying high performance computing to real applications in science and engineering and is not a benchmark exercise! The used earthquake simulation code SeisSol is a complex, real-world application, with tens of thousands of lines C/C++ and Fortran code.

Earthquakes strike suddenly without warning and the aftereffects can be disastrous. It can take lives, break bridges and cause fires! What can we do today to help prepare for earthquakes and to contain the aftereffects? While developing stronger building codes and performing earthquake drills may be part of the answer, the complete answer will be unknown until the phenomenon of earthquake rupture and its fault branching is studied thoroughly. This is exactly the problem geologists, mathematicians and computer scientists at Technische Universität München and Ludwig-Maximilians-Universität München are trying to solve with help from Intel researchers. The German scientists sought out to use the SuperMUC, Stampede and Tianhe-2 supercomputers to create high-resolution and high-order realistic 3D seismic simulations employing SeisSol.  Hence large earthquake events are complex processes including multi-physics scenarios at high frequencies, and simulating these in realistic 3D Earth models require peta- or even exascale computing power. Insights gathered from these simulations are of high relevance for scientific and industrial applications that help societies be best prepared for natural disasters such as earthquakes.

This is where Intel’s strong commitment to high performance computing and its research expertise in parallel computing comes into a synergistic play with the domain knowledge of our customers. Working in close collaboration with Intel Labs Parallel Computing Lab , performing a series of architecture-aware optimizations, the team was able to scale the complexity of science and sustained performance to an unprecedented level.  SeisSol sustained   8.6 PFLOPS (double precision), almost equivalent 8.6 quadrillion calculations per second when processing seismic wave phenomena using half of the Tianhe-2 supercomputer (only half of the system was made available to the team), implying almost 18 PFLOPS for the full Tianhe-2 machine. This amounts to the highest-ever sustained application-level performance for any supercomputing platform.  Equally noteworthy is the overall time-to-solution boost credited to the Intel® Xeon Phi ™ coprocessor used in Tianhe-2 system. The SuperMUC and the Tinahe-2 supercomputers have comparable size: 8192 nodes of Intel® Xeon® with Intel® Xeon Phi ™ on Tianhe-2 versus 9216 nodes of Intel® Xeon® on SuperMUC. About 8x more peak performance of the Tianhe-2 machine improved overall time-to-solution for the 1992 Landers earthquake simulation scenario by about 2.7x!

In a pioneering simulation of the 1992 7.3M Landers earthquake the mentioned team of scientist was able to achieve a highly detailed rupture evolution and ground motion at frequencies up to 10 Hz. Simulating this particular earthquake with high resolution (200 m fault resolution) required processing 191,089,540 tetrahedrons (Figure 1) over hundreds of thousands of time steps. Figure 2 shows how the seismic waves are emitted by the complex fault geometry into the earth resulting into ground motions which are hazardous.  Especially in case of the 1992 Landers earthquake an accurate simulation of this faulting process results into a challenging multi-physics process including fault-branching and rupture jumps between fault branches.

fault

Figure 1: 200m Fault resolution and the unstructured mesh containing nearly 200M cells.

 

landers

Figure 2: Seismic waves which are emitted by the complex fault system of the 1992 Landers earthquake scenario

 

While the excitement will continue until the winner is announced in three months, Intel and its partners are feeling proud and privileged to be nominated for this honorable award.

Figures taken from the Full Landers earthquake simulation video that can be viewed here.

 

 

Divya Kolar

About Divya Kolar

Divya Kolar holds a M.S in Computer Science conferred in 2006 from Portland State University. She joined Intel in 2005 and has previously worked as a Software Engineer where she was an active researcher in various security and manageability technologies like Intel® Active Management Technology. Today she is a Vision Strategist in the Intel’s largest research group and is responsible to promote Intel technologies to external media partners besides performing ecosystem enabling and competitive technology analysis for Intel Labs’ microprocessor research. Besides her responsibilities at Intel she has always been enthusiastic in promoting and encouraging young adults to stay in computing. She is an active board member for the largest women employee group at Intel and has been an active member in Anita Borg Institute and local SWE chapters since 2007 and has conducted multiple presentations at these conferences for over 5 years.

Comments are closed.