A Constrained, Possibilistic Logical Approach for Software System Survivability Evaluation

Abstract

In this paper, we present a logical framework to facilitate users in assessing a software system in terms of the required survivability features. Survivability evaluation is essential in linking foreign software components to an existing system or obtaining software systems from external sources. It is important to make sure that any foreign components/systems will not compromise the current system's survivability properties. Given the increasing large scope and complexity of modern software systems, there is a need for an evaluation framework to accommodate uncertain, vague, or even ill-known knowledge for a robust evaluation based on multi-dimensional criteria. Our framework incorporates user-defined constrains on survivability requirements. Necessity-based possibilistic uncertainty and user survivability requirement constraints are effectively linked to logic reasoning. A proof-of-concept system has been developed to validate the proposed approach. To our best knowledge, our work is the first attempt to incorporate vague, imprecise information into software system survivability evaluation.