A control-theoretic approach to flow control
SIGCOMM '91 Proceedings of the conference on Communications architecture & protocols
A measurement-based admission control algorithm for integrated services packet networks
SIGCOMM '95 Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Cluster reserves: a mechanism for resource management in cluster-based network servers
Proceedings of the 2000 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Service Accelerator (SEA) System for Supplying Demand Oriented Information Services
ISADS '97 Proceedings of the 3rd International Symposium on Autonomous Decentralized Systems
Mobile software agents: an overview
IEEE Communications Magazine
Web server support for tiered services
IEEE Network: The Magazine of Global Internetworking
Hi-index | 0.00 |
Due to the advent of electronic commerce, the service provider can offer the customer a Service Level Agreement (SLA) to distinguish itself in today's competitive market. Different users require different levels of service in terms of information consumption or response time, appearing a need of a system able to assure the heterogeneous service levels of the users in such a mission-critical environment. In this paper we propose an information service infrastructure based on the Faded Information Field (FIF), a distributed service replica network founded on Push/Pull mobile agent technology. The cooperation between the autonomous agents and the nodes allows the online maintenance and auto-reconstruction of the FIF system in order to satisfy the different timely requirements of each user even in an ever-changing environment. Moreover, the system is able to provide different maximum information volume according to the user's service level, a double-sided feature that benefits to both the provider's business strategy and the differentiated timeliness access technology presented on this paper. The effectiveness of the proposed technology has been proved through evaluation and the simulation results show that the system is able to adapt and satisfy the different requirements of timeliness and the different levels of information volume required by each user under changing workloads, compared with conventional systems.