Earthquake engineers have traditionally investigated the behavior of structures with either computational simulations or physical experiments. Recently, a new hybrid approach has been proposed that allows tests to be decomposed into independent substructures that can be located at different test facilities, tested separately, and integrated via a computational simulation. We describe a grid-based architecture for performing such novel distributed hybrid computational/physical experiments. We discuss the requirements that underlie this extremely challenging application of grid technologies, describe our architecture and implementation, and discuss our experiences with the application of this architecture within an unprecedented earthquake engineering test that coupled large-scale physical experiments in Illinois and Colorado with a computational simulation. Our results point to the remarkable impacts that grid technologies can have on the practice of engineering, and also contribute to our understanding of how to build and deploy effective grid applications.

A project supported by the PEER-Lifelines Program is currently finalizing a pilot system architecture supporting web-based interfaces capable of providing access to publicly available geotechnical data sets. This paper describes the pilot system architecture design developed by the project's Virtual Data Center Work Group (VDCWG). The design is based on the results of a geotechnical data use and user scenario survey and a data dictionary developed by the project's User Scenario Work Group (USWG) and Data Dictionary Work Group (DDWG) respectively. The design allows multiple data providers to make their data available through a uniform web interface, while each retains possession and control of their data. A user is able to view and download data from multiple organizations in a uniform file format. Over the long-term, the main objective is to extend the pilot system and link multiple databases of government agencies, universities and private companies. A workshop planned for June 21 - 23, 2004, will review the system to obtain input and consensus from the geotechnical community. The workshop proceedings will serve as a future implementation plan for expansion into a large-scale web portal.

Abstract: this paper, variousNEESgrid components are introduced and their roles in NEESgrid are identified. Subsequently, the MultiSiteOnline Simulation Testbed (MOST) experiment, which was conducted on July 30, 2003 to showcasethe NEESgrid software capabilities, will be discussed. The MOST experiment coupled two large-scalephysical experiments in Illinois and Colorado with a computational simulation. This paper also describesthe Mini-MOST experiment, which is a tool for NEESgrid education, training ...