Monitoring and control of networked systems with mobile agents: Algorithms and applications
Abstract
This dissertation presents a set of mechanisms, algorithms, and applications of mobile agents for monitoring and control of networked systems, wide-area distributed systems that consist of autonomous computing nodes interconnected via high-speed communication channels. The research reported in this dissertation is motivated by the necessity to develop networked systems in complex environments within which we need to separate control and monitoring tasks from the functional aspects of individual components of the networked systems. Mobile agents are generally active mobile objects with some level of intelligence that are capable of making autonomous decisions and interacting with peer agents in various ways. Their mobility and autonomy make them suitable for controlling the components of the networked systems, whereas the components are only responsible for performing given tasks. This approach simplifies the design and implementation of the networked systems; developers can concentrate only on the functional dimension of component development, while the control and communication of components will be managed by mobile agents. This dissertation makes contributions in three areas: (i) the functional extensions of an existing agent framework, (ii) the distributed and adaptive algorithms to detect agent failures and disseminate the fault information, and (iii) applications of mobile agents to Web benchmarking and a Web server that supports differentiated services, as well as to adaptive MPEG streaming and the resource directory, presented in that order.
Degree
Ph.D.
Advisors
Marinescu, Purdue University.
Subject Area
Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.