iNet-EGT: An Evolutionarily Stable Adaptation Framework for Network Applications

This paper studies a bio-inspired framework, iNet-EGT, to build autonomous adaptive network applications. In iNet-EGT, each application is designed as a set of agents, each of which provides a functional service and possesses biological behaviors such as migration, replication and death. iNet-EGT implements an adaptive behavior selection mechanism for agents. It is designed after an immune process that produces specific antibodies to antigens (e.g., viruses) for eliminating them. iNetEGT models a set of network conditions (e.g., workload and resource availability) as an antigen and an agent behavior as an antibody. iNetEGT allows each agent to autonomously sense its surrounding network conditions (an antigen) and select a behavior (an antibody) according to the conditions. This behavior selection process is modeled as a series of evolutionary games among behaviors. It is theoretically proved to converge to an evolutionarily stable (ES) equilibrium; a specific (i.e., ES) behavior is always selected as the most rational behavior against a particular set of network conditions. This means that iNet-EGT allows every agent to always perform behaviors in a rational and adaptive manner. Simulation results verify this; agents invoke rational (i.e., ES) behaviors and adapt their performance to dynamic network conditions.