Resource management using multiple feedback loops in soft real-time distributed object systems

This paper describes a Resource Management System for a soft real-time distributed object system that is based on a three-level feedback loop. The Resource Management System employs a profiling algorithm that monitors the usage of the resources, a least laxity scheduling algorithm that schedules the methods of the tasks, and hot spot and cooling algorithms that allocate and migrate objects to balance the loads on the resources. The Resource Management System consists of a single Resource Manager for the distributed system, and a Profiler and a Scheduler located on each of the processors in the distributed system.