at the University of Oregon's Department of Computer and Information Science
Keywords: Parallel and distributed computing, resource management, allocation, mapping, task assignment, scheduling, distributed shared memory.
Resource management is a key area of research in the drive to fully realize the performance potential of parallel and distributed computing systems. The complexities involved in managing both hardware and software resources that number in the hundreds to thousands, under the demands of a diverse multi-user workload, presents a spectrum of challenging problems for operating systems designers.
Our research group focuses on the management of processors and processes in systems ranging from message-passing multicomputers to loosely-coupled workstation-based distributed systems. We have developed algorithms and software tools for allocation, mapping, placement, scheduling and migration, with extensions to support fault tolerance, heterogeneous environments, and real time constraints. We have also begun to explore issues related to parallel I/O allocation and scheduling. Our approach involves a blend of theoretical, experimental, and systems development work.
Group members are sporting their ProcSimity t-shirts.