JOVE: A Dynamic Load Balancing Framework for Adaptive Computations on an SP-2 Distributed-Memory Multiprocessor

Computational requirements of full scale computational fluid dynamics change as computation progresses on a parallel machine. The change in computational intensity causes workload imbalance of processors, which in turn requires a large amount of data movement at runtime. If parallel CFD is to be successful on a parallel or massively parallel machine, balancing of the runtime load is indispensable. Here a framework is presented for dynamic load balancing for CFD applications, called Jove. One processor is designated as a decision maker Jove while others are assigned to computational fluid dynamics. Processors running CFD send flags to Jove in a predetermined number of iterations to initiate load balancing. Jove starts working on load balancing while other processors continue working with the current data and load distribution. Jove goes through several steps to decide if the new data should be taken, including preliminary evaluate, partition, processor reassignment, cost evaluation, and decision. Jove running on a single IBM SP2 node has been completely implemented. Preliminary experimental results show that the Jove approach to dynamic load balancing can be effective for full scale grid partitioning on the target machine IBM SP2. 1. Dept. of Computer and Information Science, New Jersey Institute of Technology, Newark, New Jersey 07102-1982, [email protected]. This work was supported in part by the NASA/University Joint Venture in Research Program. 2. NAS Applied Research Branch, Mail Stop T27A-1, NASA Ames Research Center, Moffett Field, CA 94035-1000, [email protected]. The author is an employee of Computer Sciences Corporation. This work is supported through NASA contract NAS2-12961.