Utilization-based heuristics for statically mapping real-time applications onto the HiPer-D heterogeneous computing system

Real-time applications continue to increase in importance as they are employed in various critical areas, such as command and control systems. These applications have traditionally required custom-made systems to execute them. Recently, with the widespread use of increasingly powerful commercial off-the-shelf (COTS) products, some real-time system designers have started a shift from custom development to COTS-based systems to achieve lower costs and more flexible systems. This research investigates the problem of allocating real-time applications to a set of COTS heterogeneous machines connected together by a COTS high-speed network. The heuristics were implemented on the High Performance Distributed Computing Program's (HiPer-D) Naval Surface Warfare Center (NSWC) testbed. At the specification of NSWC, the goal of this study is to design static resource allocation heuristics that balance the utilization of the computation and network resources in the HiPer-D system based on the system information provided. The broader goal is to maximize the time before dynamic reallocation is required for managing an increased workload at runtime.