A Parallel Workload Model and its Implications for Processor Allocation
Abstract: We develop a workload model based on the observed behavior of parallel computers at the San Diego Supercomputer Center and the Cornell Theory Center. This model gives us insight into the performance of strategies for scheduling malleable jobs on space-sharing parallel computers. We find that Adaptive Static Partitioning (ASP), which has been reported to work well for other workloads, is inferior to some FIFO strategies that adapt better to system load. The best of the strategies we consider is one that explicitly restricts cluster sizes when load is high (a variation of Sevcik's A+ strategy.