Building Self-Configuring Data Centers with Cross Layer Coevolution
Paskorn Champrasert, Junichi Suzuki · Journal of Software · 2007
Abstract—This paper describes a biologically-inspired archi-tecture, called SymbioticSphere, which allows data centers to autonomously adapt to dynamic environmental changes. SymbioticSphere follows biological principles such as decen-tralization, evolution and symbiosis to design application services and middleware platforms in a data center. Each service and platform is designed as a biological entity, and implements biological behaviors such as energy exchange, migration, reproduction and death. Each service/platform also possesses behavior policies, as genes, each of which de-fines when to and how to invoke a particular behavior. This paper presents a set of behaviors for services and platforms, and describes how services and platforms act and interact with each other. Simulation results show that services and platforms autonomously adapt to dynamic network condi-tions (e.g., user location, network traffic and resource avail-ability) by evolving their behavior policies across genera-tions. Simulation results also show that services and plat-forms coevolve to improve their adaptability by adjusting their behavior policies cooperatively. Index Terms—Autonomic self-configuring network systems, Biologically-inspired networking, evolvable network systems I.