An adaptive network policy management framework based on classical conditioning

With the future vision of autonomic computing, policy-based management becomes a promising solution, but its static policy configurations can not accord with the target of self-management. Inspired by classical conditioning, the basic learning mode of biological system, we presented a dynamic policy adaptation framework, which is founded with several simple building blocks composing a complete reflex arc, and it is an extension of Internet Engineering Task Force (IETF) framework for policy-based management. In order to learn the most suitable configuration policy from system behavior, network rules specified within our framework are dynamically triggered by comparing training stimulus of the experiments. Some typical examples are introduced to testify how network management policies cater for the dynamic management of network security, and the selected network security policy is varied with changing environment at run-time. Our approach provides the flexibility of adapting to the changed network environment and the ability of simulating some typical experiments of classical conditioning. Furthermore, the major advantage of this procedure is that the framework could successfully realize the self-learning process of classical conditioning and achieve an adaptive network policy management.