Title:	TBAGENT-BASED FRAMEWORK FOR ADAPTIVE POLICY MANAGEMENT IN AUTONOMIC COMPUTING SYSTEMS
Author(s):	Karthikeyan Ponnalagu , Nanjangud C Narendra
ISBN:	972-8924-09-7
Editors:	Nuno Guimarães, Pedro Isaías and Ambrosio Goikoetxea
Year:	2006
Edition:	Single
Keywords:	Agents, Aspect Oriented Programming, Autonomic Computing, Policy Management.
Type:	Full Paper
First Page:	297
Last Page:	306
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Current software systems contain static configuration models implemented as a combination of editable configuration documents in formats such as XML and the corresponding business logic distributed across the systems. There is no explicit understanding recognized on making those choices in the configuration documents. Similarly the corresponding system behaviour specific to the configuration values are contained implicit, making the system very rigid in nature. Thus the system behaviour is typically controlled by implicit functional policies defined and enforced at build time. With enabled interoperability between disparate systems implemented in this manner, managing the resources and behaviour within such systems has become increasingly difficult. One promising approach to address this challenge is to explicitly define the controlling policies and enforce them with a centralized management system. Policies are considerations designed to guide decisions on courses of action and can be used for numerous purposes within configurable systems. Throughout this paper, we take the example of autonomic systems that follow enforced policies for incorporating a set of decision-making technologies into its management components in order to simplify and automate the administration of the functional components of the system. There is a possibility that such a system could fail due to inconsistent or unrealizable application of the policies in changing contexts. In this paper, we describe an adaptable agent based run time framework for preventing the system failures due to such inconsistent policies. This framework is based on aspect-oriented programming principles for providing a flexible and efficient way to inject new functional policies and validate policy conflict resolutions without the need for extensive re-programming of the autonomic system.
	Go Back