Reducing the Communication Overhead of an Artificial Hormone System for Task Allocation by a Task Window

The Artificial Hormone System (AHS) is a completely decentralized operation principle for a middleware which can be used to allocate tasks in a system of heterogeneous processing elements (PEs) or cores. Tasks are scheduled according to their suitability for the heterogeneous PEs, the current PE load and task relationships. The AHS also provides properties like self-configuration, self-optimization and self-healing by task allocation. The AHS is able to guarantee realtime bounds for such self-X-properties.

If a large number of PEs applies for a large number of tasks a considerable amount of hormone communication is produced to assign these tasks to PEs. Until now this could be only circumvented by limiting the number of tasks a single PE applies for. However, this reduces the self-optimization and self-healing properties because the number of possible PEs to execute a tasks is limited in the same way. In this paper we present the concept of the task window as a new approach to significantly reduce the communication overhead without affecting self-optimization and self-healing properties. We additionally show that the task window can even improve the real-time behavior of task allocation.