An Architecture Of Multiagent System For Autonomic Computing And Its Actualization

Autonomic computing is a concept appearing recent years. The objective of Autonomic computing is to enable applications, systems, and entire networks to become more self-managing. Self-management involves four qualities--self-configure, self-heal, self-optimize, and self-protect. Autonomic computing is one of the technologies that enable the creation of an information technology computing infrastructure for the next era of computing. The rise and development of Autonomic Computing is resulted in the following three reasons:1.the increase of software's scale and complexity,that result in the increase of pressure of management and maintenance;2.the increase of requestion for the robustness of software,that may decrease the cost of mistake.3.the gap between released software and runtime software,that require software increase the quality of service.At first this paper introduce the concept and method of Autonomic Computing, analyze the architecture and the essential technology for Autonomic Computing.Then, this paper propose an architecture for Autonomic Computing ASOAC following the FIPA standard.ASOAC base on the Mutiagent system,it's used for the design and implement of Autonomic systems,it's an open and extensible architecture.The Autonomic Manager in the ASOAC is actualized according the AM-ASOAC architecture proposed in this paper.The ASOAC provides common service for the Autonomic Manager,and the Autonomic Manager in it can perform function such as negotiation, collaboration and reasoning they can also communion with each other.The ASOAC is carried out in an oil ensurence system,the result of it prove that systems using the ASOAC have ability of highly self-managing,and the ASOAC is a feasible architecture for the Autonomic resolving of complex cooperative problem in a changing environment.