Multi-Granularity Dynamic Characteristic Analysis Of Networked Software

Since the nineties of the last century, the Internet accesses business applications. With the rapid development of computers and communication technologies, the Internet has become the important information infrastructures for advancing the world economic development and social progress. With the continuous emergence of cloud computing, peer-to-peer computing, pervasive computing, and the other new emerging computing model, the Internet led to a revolution in the software industry-software networks. It promotes the software engineering to quick access to the networked era.In the complex networks environment, networked software can realize the interconnection, interoperability, interoperability of network resources via the interaction and collaboration between various components of networked software as well as between various components of networked software and their environment. With the integration between software and network, the complexity of networked software increases with the explosive exponential growth. In the complex network environment, as the users in the role, needs goals, business processes and service resources, in different degrees of difference, the users cannot explain their demand exactly, clearly, and completely. To take on the above difficulties and challenges, it is necessary to develop a new approach for the modeling, analysis, and evolution of networked software.By using the fundamental theories and approaches of complex systems and complex networks, this paper further investigates the modeling, analysis and evolution of networked software. Moreover, the above theoretical results are applied to the development and platform construction of networked e-commerce software. Main results are summarized as follows:Based on the structural features and analysis of spatial and temporal scales, we propose the dynamical systems modeling method of networked software based on state space and complex adaptive systems modeling method of networked software. From the micro to the macro network resources modules of different granularity, we further explore the origin of complexity of networked software, self-organization and emergent phenomena.Based on the theory of complex adaptive systems, we introduce the scaling theory of networked software and also apply the above scaling theory to analyze the dynamical behaviors of networked software. Furthermore, we reveal the dynamic relationship between the topological structure and function of networked software, the dynamic relationship between the community structure (or module) and function of networked software, the multi-scale analysis, the dynamic interaction among the various components, the dynamic interaction between the various components and their environment, the monitoring and evaluation of dynamic characteristics, the intervention and control of dynamic characteristics, the self-organization and emergence of network collective behaviors, and the other basic features.Based on the theory of complex adaptive systems, we propose the multi-granularity dynamic evolution modeling approach of networked software, further investigate the dynamic evolution of networked software, reveal the on-demand preference dynamic evolution of networked software, adaptive evolution, on-demand control, dynamic evolution of network resources and mining, pinning control laws.Based on the above modeling and analysis of networked software, we have further analyzed the multi-granularity dynamic characteristics of Hermes MSH. Also, we propose the computing model, development model, and evolutionary model for the networked e-commerce software, reveal the open source, open, on-demand services, on-demand evolution, collaboration and interoperability, security and the other internal laws for networked e-commerce software. Moreover, we develop the networked e-commerce platform based on the cloud computing.