Modeling And Simulation Of Complex Adaptive Service System (CASS) Based On Cloud Computing

In recent years, as an emerging technology and business model, cloud computing brings a new revolution in the field of information science and technology and becomes a hot research direction in academia and industry. However, unlike the previous calculation structure and system architecture, cloud computing is based on the theory of service-oriented architecture (SOA). This new generation of computing and business models takes abstract service as the principal form which is encapsulated by virtual resources distributing on the Internet. With cloud computing characteristics, like large scale, virtualization, scalability, on-demand services and generality, have been excavated, the encapsulated resources of the cloud computing service has also changed from the initial computing resources to all other social resources, such as manufacturing resources, financial resources, etc. Therefore, the research on the service system should not be confined to the field of computational science but need integrate different research fields, such as economics, management, sociology and so on.Based on the nature of generalization of services and development trend of cloud computing, the present paper analyses the seven characteristics of Complex Adaptive System (CAS) in cloud computing and the complexity, adaptability, subjectivity of its service in combination with System Science, Computing Science, Economics, Biology and other related domain knowledge. This paper proposes a new theory:Complex Adaptive Service-based System (CASS). After this, the paper improves the simple genetic algorithm (SGA) to build the model of CASS. The basic framework of CASS model is includes coding, initial population selection and fitness evaluation. The evolutionary process of it is divided into six steps:selection, crossover, mutation, aggregation, internal and external environment and termination. Finally, this paper set up a simulation model of CASS on the NetLogo platform. Through a large number of experiments and data analysis, it verifies that the process of CASS evolution reflects seven characteristics of CAS and analyzes the influence of different behaviors and environment on the system evolution in CASS Model. To sum up, the present paper puts forward the concept of CASS, completes the modeling and simulation and preliminary analyzes the complexity and evolution characteristics of it which has a strong guidance and predict on the research and management of relevant real systems.