B. K. Alunkal, I. Veljkovic, G. von Laszewski, and K. Amin, "Reputation-Based Grid Resource Selection," Preprint ANL/MCS-P1127-1203, December 2003. [pdf]
The Grid approach provides the ability to access, utilize, and manage a variety of heterogeneous resources in virtual organizations across multiple domains and institutions. Selecting appropriate resources within such a distributed Grid environment to satisfy quality of service requirements is a complex and difficult task. This paper proposes a reputation management framework for Grids to facilitate a distributed and efficient mechanism for resource selection. Our reputation management service is based on the concept of dynamic trust and reputation adaptation based on community experiences to classify, select, and tune the allocation of entities, including resources, service, and services provided by people. The framework can evaluate through specialized services simple contextual quality statements in order to effect the reputation for a monitored resource. The proposed reputation service uses a novel algorithm for evaluating Grid reputation by combining two known concepts (a) using eigenvectors to compute reputation and (b) integrating global trust. We use the resulting framework GridEigenTrust framework.