MULTIPLE-GOAL OBJECTIVE FUNCTIONS FOR OPTIMIZATION OF TASK ASSIGNMENTIN COMPLEX COMPUTER-SYSTEMS

Complex systems are large applications, typically running on distribut ed, heterogeneous networks, driven by a number of distinct constraints and desiderata on goals such as performance, real-time behavior, and fault tolerance. These requirements frequently conflict, and satisfact ion of these design objectives interacts strongly with assignment of s ystem tasks to processors, The NSWC design framework DESTINATION provi des an assignment module which can be used to optimize the system, as measured by the value of a weighted combination of objective cost func tions, For even modest-sized systems and networks, assignment space is too large to search exhaustively; however, there are numerous algorit hms which generate heuristically good assignments. However, compile-ti me evaluation of many interesting design factors, even those clearly r elated to assignment, is impossible without some estimate of the sched ule. This paper therefore discusses approaches for determining a reaso nable ''pseudo-schedule'' for a given system, network, and assignment, and the use of this to simulate execution in evaluating cost function s.