PRIORITY ASSIGNMENT IN A NETWORK OF COMPUTERS

This research is aimed at developing analytical tools for system modeling and analysis of real-time computer networks. We formulate an idealized mathematical model for a multiserver system with a finite length nonpreemptive priority queue. Given that jobs consist of dependent tasks having linear loss functions, we formulate an algorithm for assigning priorities to tasks. We define a feasible successor set of a task as a subset of tasks which can be scheduled independently of other similar sets. task weighted by the maximum cost rate per task over every feasible successor set of the task and the task set is divided into levels based on the precedence relation among tasks. Then, generally, the algorithm gives priority to tasks within a given level according to monotonic nonincreasing weighting values. We show that the algorithm minimizes the total cost over the task set for the single computer case and the case when the task set has a tree structure. A simple counterexample illustrates that the algorithm is, in general, suboptimal. Copyright © 1969 by The Institute of Electrical and Electronics Engineers, Inc.