Brittleness analysis of software architecture based on ant colony optimization

Abstract

The term software architecture (SA) intuitively denotes the high level structures of a software system. It can be defined as the set of structures needed to reason about the software system, which comprise the software components and connectors, their relations and properties. Nowadays SA has become an important factor in the process of software development and the researches mainly focus on the languages, modeling, dynamic evolution process, etc. In this article, the complex system and brittleness theory are applied into the field of SA, and the concept of brittleness graph and collapse path of SA is introduced into the analytical process. Ant colony algorithm is used for simulation. The results of simulation demonstrate that Ant colony optimization (ACO) performs well on finding out the max collapse route of the brittleness graph of SA.