SA.Resilience.Evaluator: Evaluate the Resilience of Software Architecture Based on Attack Surface Analysis

View at publisher
Authors: Can Zhao, Jiaxin Pan, Bixin Li
Cited by: 0
References: 25

Abstract

Resilience is vital to software systems and helps them maintain an acceptable level of service in the face of various attacks. However, existing methods for evaluating the resilience of software system mainly focus on implemented software system, there is no effective method to evaluate resilience at design time. In addition, existing methods only concern specific types of software systems, lacking comprehensive consideration of multi-dimensional characteristics of software systems and they often have stringent conditions when evaluated in practical applications etc. In this paper, we propose a method for evaluating the resilience of software architecture based on attack surface analysis and it is applicable to various types of software systems. Our method first identifies potential security threats and vulnerabilities in software architecture through attack surface analysis technology. Then, our method analyzes the types and severity of potential external attacks on the software architecture by combining the CAPEC software attack database. After that, our method comes to evaluate the resilience of software architecture according to the safety, reliability, and restoration of software architecture. For the convenience of conducting the case studies, we first design and implement a tool called SA.Resilience.Evaluator and then we use the tool to evaluate the effectiveness of our method in two scenarios. In the first scenario, we use the UML activity diagram of software architecture model as the input of our tool, which was obtained from architecture designer; in the second scenario, we use architecture dependency graph as the input, which was obtained from architecture recovery techniques. The results of studies show that our method is better than existing methods in both effectiveness and efficiency.

Fields of Study & Subjects

Advanced Software Engineering Methodologies Information and Cyber Security Software System Performance and Reliability

Publication details

Journal / source
IEEE Transactions on Reliability
Volume
75
Pages
1035–1049
Publication date
1 January 2026
Language
EN