Extracting core nodes in large-scale software

Understanding the structure of software systems can potentially help the development and maintenance of a large-scale software. Some research show that a large-scale software system exhibits the similar features of a complex network. And some nodes play central role in the system. In this paper, our goal is extracting core nodes from a large, code-given software system. First, we integrate some index characterizing the status of a node, which is proposed by other researchers. However, considering the mutative operation of a running system, just only analyzing the feature of source code is not enough. Then, we put the dynamic features into our consideration. By comprehensive analysis from static and dynamic, we present our approach to obtain core nodes in a large system. At last, we adopt simulation method with NetLogo. Simulation results illustrate that our method is effective in extracting core nodes. Software engineering may benefit from our study, including system reliability and program comprehension.