Modeling software faults propagation

College of Computer Science and Technology, Jilin University; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University - Changchun 130012, PR C

^a wangjian591@gmail.com
^b lyh_lb_lk@yahoo.com.cn

Received: 3 October 2010
Accepted: 10 December 2010

Abstract

We investigate the problem of a widespread occurrence of a software fault in complex software. We propose a topological model in which directional edges and the corresponding weight values are used to represent calling relationships and close degree among functions. By introducing the function fault-tolerant capability and the software fault intensity, we design a software fault propagation model to explore interactions between fault propagation dynamics and software parameters, such as trigger methods, fault intensities, initial fault number, and fault-tolerant capability. The simulations show that real software networks exhibit much stronger fault-tolerant capability than random networks and scale-free networks. We also find out three reasons why some functions are immune against faults: the strong fault-tolerant capability of i) the node itself, ii) the neighbor nodes, and iii) the nodes with zero in-degree.