MRI: Acquisition of a Research Infrastructure for Collaborative Scientific and Engineering Computation

George Washington University, Washington DC

Investigators

Abstract

This project, acquiring four heterogeneous computing clusters connected by high-speed networks, aims at serving as the cornerstone for a Bioinformatics program. Major activities include: Promoting research in bioinformatics, facilitating exchange among scientists, Training in cluster administration leading to a certificate, and Supporting undergraduate senior design projects. The infrastructure services the following research areas: Network Algorithms and Distributed Systems, Large Scale Flow Simulations, Bio-informatics, and Distributed Data Storage for Biological Data. The first implements and evaluates network protocols and distributed systems to understand system behavior in simulation. The second uses equipment for scalable Cartesian grid methods and for determining biological flaws. The third, involving computational protein modeling, proposes algorithmic improvements to enhance the backbone of the selection in the protocol. Computational biology (phylogenetics and molecular dynamics of biomolecules) will also benefit from the cluster. The last deals with the genomics core facility. The project addresses managing, processing, interpreting, and efficient availability of data. The computing clusters, called Lion, Rhino, Sparrow, and Gazelle, are used for production, development, and training. The project emphasizes recruitment and retention of students within a strong educational component. Four new courses will be developed and a training program will award a cluster computing certificate upon completion.