Autonomic Computing, International Conference on (2007)
Jacksonville, Florida, USA
June 11, 2007 to June 15, 2007
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICAC.2007.31
Shinji Kikuchi , Fujitsu Laboratories Limited, Japan
Satoshi Tsuchiya , Fujitsu Laboratories Limited, Japan
Motomitsu Adachi , Fujitsu Laboratories Limited, Japan
Tsuneo Katsuyama , Fujitsu Laboratories Limited, Japan
Policy-based management is drawing attention as a solution to managing today?s complex information systems. To be dependable, a policy-based system must be able to check the validity of a policy written by administrators. However, common test methods such as operations tests in a test scenario and simulations cannot check whether systems with given policies will work properly in every possible situation. To solve this problem, we propose a policy verification and validation framework based on model checking that exhaustively verifies a policy?s validity by considering the relations between system characteristics and policies. We first define the validity of policies and the information needed to verify them from the viewpoint of model checking. We then construct our policy verification framework based on the definition and, finally, present a case study applying this framework to an on-demand data center scenario and show the effectiveness of our approach.
policy-based management; policy verification and validation; model checking; autonomic system; on-demand data center; resource management
S. Tsuchiya, M. Adachi, T. Katsuyama and S. Kikuchi, "Policy Verification and Validation Framework Based on Model Checking Approach," 2007 International Conference on Autonomic Computing(ICAC), Jacksonville, FL, 2007, pp. 1.