Overdrive Controllers for Distributed Scientific Computation

Abstract

Large distributed computer systems have been successfully employed to solve modern scientific problems that were previously impracticable. Tools exist to bind together notably underutilized high speed computers and sizable storage resources, creating architectures that provide performance and capacity that scales with the quantity of resources invested. In various forms such as Internet computing, desktop grids, and even "big iron " grids, users and administrators find it difficult, however, to obtain aggregate systems that are more reliable than their underlying components and simple to utilize in concert. In this work, we will propose a model for controlling complex distributed systems and its application to the construction of scientific repositories.