Towards Multi-Agent Autonomic Architectures

G. Pour (USA)

Keywords

Autonomic architecture, multi-agent architecture, agent components, space exploration, pervasive computing

Abstract

At the dawn of pervasive computing, autonomic computing is emerging as a viable approach to address the challenge of ever-increasing complexity in development, deployment, and management of computing systems. The goal of autonomic computing is to create new systems that are able to self-manage. This requires such systems to have the ability to self-configure, self-optimize, self protect, and self-heal. Autonomic computing is in its infancy. Development of autonomic architectures remains an open research issue. Our research, co-sponsored by NASA, IBM, and Hewlett-Packard, aims at addressing this issue. We have developed a multi-agent approach to designing autonomic architectures. We have also used the approach to design multi-agent autonomic architecture. In a collaborative, multi-disciplinary research project with NASA, we are exploring how to adapt our new multi agent architecture for space exploration operations. This paper presents the multi-agent approach to developing autonomic architectures. It also discusses our latest research results and our plans for the next phase of this research project.

Important Links:



Go Back