A novel approach to quality of service control in an active service network (application layer active network) is described. The approach makes use of a distributed genetic algorithm based on the unique methods that bacteria use to transfer and share genetic material. We have used this algorithm in the design of a robust adaptive control system for the active nodes in an active service network. The system has been simulated and results show that it can offer clear differentiation of active services. The algorithm places the right software, at the right place, in the right proportions; allows different time dependencies to be satisfied and simple payment related increases in performance.
|Number of pages||13|
|Journal||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Publication status||Published - 2000|