Geoinformatics 2007 Conference (17–18 May 2007)

Paper No. 6
Presentation Time: 2:30 PM-4:30 PM

MANAGING A PARAMETER SWEEP FOR EARTHQUAKE SIMULATION RESEARCH


YOUN, Choonhan1, KAISER, Tim1 and SEBER, Dogan2, (1)San Diego Supercomputer Center, University of California, San Diego, 9500 Gilman Dr. #0505, La jolla, CA 92093-0505, (2)San Diego Supercomputer Center, University of California, San Diego, 9500 Gilman Drive, Mail Code 0505, La Jolla, CA 92093-0505, cyoun@sdsc.edu

A Grid middleware consists of a set of tools and technologies that allows users to access to Grid resources and applications using a common set of protocols and services. It also provides seamless computing and information environments for science and engineering communities. Within such Grid computing environment, a Grid portal provides a user interface to a multi-tier Grid application development stack which is the Grid middleware. The GEON (GEOsciences Network) infrastructure that is naturally distributed with users and resources spread across 16 different partner sites in the US provides portal, middleware and data resources to facilitate scientific discovery using applications, tools, and services for domain scientists through a cyberinfrastructure. It consists of both service oriented Web/Grid framework and application toolkits, using Web service model and portlet programming model to represent applications. Based on those grid environments, we have developed the SYNSEIS toolkit to be a computational platform that facilitates synthetic seismogram calculations in 2-dimensional and 3-dimensional media for research and educational purposes. It is a web-based simulation and analysis system and is one of the services provided within the GEON network. SYNSEIS provides information management, computational analyses and simulations all wrapped as Web services and Grid computing. The E3D simulation software used by SYNSEIS is designed to simulate seismic wave propagation through the Earth's interior. We grid-enabled E3D using our own dialect XML inputs, running crustal models through Web services. The XML inputs for this application include: structural information which contains cell dimension, number of time steps, source parameters, geology, and number of stations. We demonstrate how one can use a simple management scheme to perform a parameter sweep and spread the work across a collection of computational resources, using an application that was not specifically designed to perform parameter sweeps. In particular, we identify the earthquake simulations in SYNSEIS as an example application that can benefit from running on multiple computational resources and subsequently promote the sharing of computational resources among partner sites involved in the GEON project.