Study On Biological Network Platform Based On Self-Adaptive Software Architecture

With the development of network application towards high-performance, large-scale, diversity, people put forward more distributed requests to the Internet: the network needs some characteristics, such as self-extension, mobility, survival, simplicity, and adaptability to the long and short change to the users and network environments. As such, it is necessary to optimize the Internet architecture and design its application.Biological information systems can be regarded as distributed automatic systems, and can provide the effective technologies and methods for science and engineering field. In biological world, the large-scale systems, such as human society, bee swarm, and biological immune systems have formed many important theories and mechanisms that can be satisfied with the future requirements of the Internet.In the thesis, we firstly show the former bio-network computing model in order to study the bio-network architecture better. The model mainly includes bio-network platform (which consists of bio-entity context, bio-network services, and bio-network components) and bio-entities.Based on the adaptive software architecture, we introduce design and implementation of building the bio-network platform in theory. We provide a communication technique to implement the distributed network component object transfer. The component on bio-model sends message to communicate with others. We study the communication mechanism whicn can coordinate among bio-entities, and between bio-entities and super-entities, and among super-entities.Applying the correlated theories and mechanisms, we design new entities — scheduling entities which are special cyber entities in our bio-network framework, and utilize scheduling entities to guide entities moving, which is very useful to the load balance in adaptive systems. The foundation of scheduling entities will be useful to the bio-network to better satisfy the adaptability of the future Internet.We present a genetic algorithm-based load balance algorithm that can achieve resource optimizations. The optimization goal is load balance of network, which makes light load computer have the ability to service. Through our algorithm, the bio-network could reasonably utilize its resources and further improve its adaptation ability. Finally, we show the simulation experiments to demonstrate the effectiveness of our approach.Through collaborations and dynamic choices of multi-agents, the bio-network accomplishes its service and applications. We provide an intelligent evaluation and optimization model for the quality of service in multi-agent system based on the integrated method of fuzzy logic, neural networks and genetic algorithms. The model is a five-layer fuzzy neural network optimized by genetic algorithm. The simulation result proves that the model can efficiently resolve the fuzzy and non-certain problem of the quality of service in multi-agent systems.Combining with other biological mechanisms, services and applications of the Internet are designed and validated through computer simulation. The behaviors of bio-entity and super-entity are simulated to produce services and applications needed.Finally, we conclude the whole thesis, and make further researches on theories and applications of the bio-network model.