An Approach of Agent-Based Architecture for Autonomic Network Management

Autonomic network management is viewed as the capability of network entities to self-government, which frees IT professionals from time-consuming and tedious configuration tasks. Hence, IT professionals can focus on high level management (business goals, for example). In this paper, we propose an approach of agent-based architecture for autonomic networking on the basis of Motorola's FOCALE architecture. By adding an adaptive mobile software agent to each managed resource, it makes the managed resources intelligence and simplifies the function of MBTL (Model-Based Translation Layer). Since we use component based technology to construct agents, the agents are custom-built for the specific resources, which is convenient, fast, reusable and adaptable. The agent-based autonomic networking architecture improves FOCALE architecture, especially in the relationship between managed resources and autonomic manager. The architecture is compatibility to legacy resources (have no intelligence), which will greatly promote the application of autonomic networking.