Geoinformatics: A Petascale Cyberfacility for Physics-Based Seismic Hazard Analysis (PetaSHA-2)

University Of Southern California, Los Angeles CA

Investigators

Abstract

This project is an outgrowth of the Community Modeling Environment (CME) that began in 2001. The PetaSHA project attempts to build on NSF's goal of having petascale computers operational by 2010. The plan is to increase the resolution of dynamic rupture simulations by an order of magnitude to gain a better understanding of the physics involved. The Principal Investigators also plan to increase the frequency of ground motion simulations above 1 Hz and to use realistic 3D structural models for Southern California. These structural models will be validated and improved using full 3D waveform tomography. All of these computationally-intensive tasks will be used to inject more physics into probabilistic seismic hazard analysis. This project (PetaSHA-2) is a follow-on to PetaSHA-1, which began in October, 2006. It will benefit from the PetaShake platform which is being funded under NSF/OCI's PetaApps program. This project is tightly coupled to the PetaShake project. A goal of this project is to scale the performance of PetaSHA-1 to petaflops, along with the addition of a few new capabilities, such as full 3D waveform tomography (F3DT) and the BroadBand delivery platform. Much of the work will be devoted to developing the infrastructure necessary to scale to petaflops.