International Journal of Computer Theory and Engineering

Abstract

Survivability represents a system’s capability to withstand malicious attacks and system failures in order to provide essential services to users even in a challenging environment. In a proof-carrying paradigm, a user publishes his/her survivability requirement policy and a system provider constructs a proof that the system satisfies the user’s requirements. Finally, the user verifies if the proof is valid. In this paper, we discuss proof approaches and techniques used by the system provider to automatically compile such a proof. We develop algorithms to show how different proof choices are generated so that the system provider can choose the most cost-efficient approach in the proof process. Proof generation relies on the certifications generated by trusted evaluators. We show the necessary steps to construct the basic proof elements which can be logically linked to form the ultimate proof.