Highly parallel computers have the memory capacity and potential speed to perform very high-resolution time-dependent calculations. Parallel computers with hundreds of fast processors require highly scalable algorithms to avoid wasting expensive resources. On these machines careful attention must be given to program design to fully exploit scalable algorithms. We have proposed a programming model that can express a class of scalable algorithms. In this paper we show how compiler analysis can ease the task of writing programs within this programming model. We have used this programming model manually and have achieved good results in reducing communication and synchronization overhead. We describe early results of a prototype compiler transformation tool called TORAZ that applies this analysis to real programs.
展开▼