Design of a multi-agent system for autonomous database administration

As more and more businesses and services find themselves online due to the Internet phenomenon, a plethora of information sources is now available. These trends have rendered databases an indispensable component of daily life and have increased their complexity significantly, which has, in turn, amplified the workload on database administrators (DBAs). This has given rise to an increasing need for self-managing and self-administering databases. We present the analysis and design of a novel and innovative solution to address the problem of overburdened and expensive DBAs. We propose a self-administering wrapper around database systems in the form of an intelligent multi-agent system tool that autonomously and rationally administers and maintains relational databases. A planned implementation of the agent-based system which proactively or reactively identifies and resolves a small sub-set of DBA tasks is discussed and the Gaia methodology is used to outline the detailed analysis and design of the same using role models and interaction models. A brief description of the functionalities, responsibilities and components of each agent in the planned multi-agent system is presented.