Toward an evaluation framework for reputation systems in autonomic computing networks

Today´s cyber-infrastructure is significantly reshaped with the advent of autonomic computing. The quality and robustness of services in such networks, however, would not be ensured and optimized without an efficient auto-defense mechanism. Reputation system, which has been widely applied in peer-to-peer networks, commercial online applications, and mobile ad hoc networks, potentially plays an important role in achieving auto-defense functionality by fully exploring and tightly integrating self-protecting and self-healing properties of autonomic computing systems. The purpose of this paper is to provide a critical evaluation framework for examining the quality and soundness of the reputation calculation models which can be possibly applied in autonomic computing networks, leading to a set of criteria for the selection of appropriate reputation systems in accordance to the working, economical and social needs of application scenarios. By exemplifying a number of representative reputation systems, we also illustrate that the framework may serve as an analytical basis for identifying and comprehending the operational limits of a reputation system, thus suggesting effective enhancements.