Dynamic Migration Algorithms for Distributed Object Systems

Complex distributed object systems require dynamic migration algorithms thnt allocate and reallocate objects to respond to changes in the load or in the availabilip of the resources. I n this paper we present the Cooling and HotSpot migration algorithms that reallocate objects when the load on a processor is high or when the latency of a task is high. The algorithms have been iniplenietited as a feedback loop in the Eternal Resource Management System where irlforniation obtained from monitoring the behavior of the objects and the usage of the processors' resources is used to dynamically balance the load on the processors and iniprove the latency of the tasks. The cost of moving an object is justiJied by urnortization over many method invocations, and constrains the rate at which objects are moved. The experinienlal results show That our algorithms guarantee steadyjlow of operation for the tusks andgracefully migrate objects from the processors when processor overloads and high task latencies are detected.