A Grid computing system is a virtual resource framework. Inside the framework, resources are being shared among autonomous domains which can be geographically distributed. One primary goal of such a virtual Grid environment is to encourage domain-to-domain interactions and to increase the confidence of domains to utilize or share resources without losing control and confidentiality. To achieve this goal, a Grid computing system can be viewed more or less as a human community and thus the “trust” notion needs to be addressed. To integrate trust into a Grid, some specific issues need to be considered. In this paper, we view trust in two aspects, identity trust and behavior trust. Further, we briefly present two important issues which help in managing, evolving and interpreting trust. The two issues are grid context and trust tree structure.
Woodas W. K. Lai, Kam-Wing Ng, Michael R. Lyu