DYNAMIC TASK ALLOCATION ON OPEN HETEROGENEOUS PROCESSOR SYSTEMS

A dynamical model for the distribution of tasks on a system of heterogeneous Processors is introduced. The task distribution procedure is based on the perceived benefits a task sees in using a particular processor for the execution of its requirements. The decision is usually based on a limited amount of information and is therefore a statistical process. The degree of certainty can be regulated by the so-called gain factor and ranges from a total lack of knowledge to a complete knowledge which renders the distribution dynamics a deterministic one. Two different entropy functions are introduced. Together they serve as a metric for the efficiency and the suitability of the processor system for the tasks that are being distributed on it. A method is developed which allows a system administrator to evaluate the optimal task distribution, subject to the limited system resources. The simulated distribution values are to be compared to these optimal values.