Adapted quality resource selection using the Grid reputation-policy trust management service

The concept of trust and reputation had a long profound impact on the way systems were measured in terms of their credibility and performance. This has consecutively lead into numerous reputation-based trust management systems being deployed for various computing environments as a decision support tool for assessing the trustworthiness of participating parties. In the context of Grid computing, reputation-based trust management systems play an important role for supporting coordinated resource sharing as they can reduce job execution failure by selecting relatively competent resources based on aggregated historical recommendations and given job requirements. In this paper, we present the Grid reputation-policy trust management service (GREPTrust) for managing resource selection in computational grids. This infrastructure level service encompasses a novel reputation-policy trust model, which enables service consumers (e.g. monitoring toolkits and resource brokers), to carry out an active participation in the trust and reputation evaluation processes. This is achieved by enabling service consumers to augment standard reputation queries with a set of reputation-policy statements rectified as trust decision strategies. Each strategy forms complete trust metrics blueprint for the reputation algorithm and therefore allows fine-grained resource selection adapted to specific job requirements.