MRI: Acquisition of an Interdisciplinary Facility for High-Performance Computing

University Of Maryland Baltimore County, Baltimore MD

Investigators

Abstract

Proposal #: CNS 08-21258 PI(s): Gobbert, Matthias K.; desJarding, Marie E.; Olano, Thomas M.; Rheingans, Penny; Sparling, Lynn Institution: University of Maryland - Baltimore County Baltimore, MD 21250-0002 Title: MRI/Acq.: Acq. of an Interdisciplinary Facility for High-Performance Computing Project Proposed: This project, creating a high-performance computing cluster with many nodes and a state-of-the-art InfiniBand interconnect, aims to form the central resource for high-performance computing dedicated to support interdisciplinary research and training across the entire campus. Supporting 23 researchers and ten departments and centers, the cluster provides an opportunity to foster an emerging community of interdisciplinary researchers interested in computational science. The facility enables the solutions of scientific problems several magnitudes larger than currently possible. The project brings together researchers from Computer Science, Mathematics, Physics, Biology, Statistics, Economics, and Engineering (Electrical, Mechanical, Civil, and Environmental). In the earth sciences, the research helps to reduce uncertainty in policy decisions about water quality in the Chesapeake Bay watershed, allows analysis of terabyte-scale data sets for understanding climate change and its natural anthropogenic origins, and provides insights into the predictability of hurricane intensity. In Engineering, the research helps to assess the reliability of short-pulse laser systems in communication systems. Biomedical research tries to establish the links between protein structure and disease, and also directions in the design of biomaterials. Applications in geodesy and geomagnetic data assimilation improve understanding of the Earth?s gravitational field and basic processes within the Earth?s core. The basic research in visualization of complex systems, the development of efficient parallel algorithms, and the utilization of graphics hardware for scientific computing all have a broad range of practical applications. Broader Impacts: Improving research collaborations, the instrumentation should lead to scientific and technical advances, have deep impact in education and research training of all users, and contribute to workforce training. Moreover, the increased collaboration and communication makes it easier to involve undergraduates and underrepresented groups in the area in this undergraduate and minority-serving institution.