Research On Dispersed Computing Architecture And Its Optimization Based On Storage,Computation And Communication

In recent years,with the rapid development of information technology,people's pursuit of quality of life has continuously improved,and interconnected applications in various vertical industries have also emerged,such as autonomous driving,smart home and smart medical care.These emerging applications have different requirements in terms of storage,computation and communication,which poses a huge challenge to the current traditional computing and network architecture.Therefore,how to design and use a dispersed computing architecture that integrates storage,computation and communication to adapt to the digital and intelligent development of various vertical industries has become a research hotspot in the academic world.Although many researchers have carried out some research work on the above issues,there is still room for improvement in the following three aspects of the current research: 1)Existing architectures such as mobile cloud computing and mobile computing will generate network congestion and large system energy consumption,which is difficult to provide strong integrated service capabilities for these emerging interconnected applications.2)With the increase in the amount of application data and task requests in various industries,there will be some competition for resources when the system integrates storage,computation and communication,making it difficult for the system to guarantee a better performance level.At the same time,wireless terminal equipments still have certain energy consumption constraints,which will make it difficult to maintain the ecological stability of the entire computing system.3)Although the industry has developed corresponding simulation platforms for cloud computing,there is still a lack of certain simulation platforms and prototype systems for a dispersed computing architecture that integrates storage,computation and communication.To this end,this paper studies the dispersed computing architecture based on the integration of storage,computation and communication,and its optimization technology.The paper mainly performs the research from the aspects of computation architecture,key technology for resource optimization and system platform verification,providing part of the prototype for future computation and communication frameworks.The research contents and contributions of this paper are summarized as follows.1.System design and analysis of dispersed computing architecture.Aiming at the shortcomings of the existing mobile computing architecture,this paper proposes a cloud-edge-end dispersed computing architecture that integrates storage,computation and communication based on the current research.The system architecture is generally divided into cloud layer,edge layer and terminal layer.The top cloud layer is mainly the global cloud data center.The middle edge layer mainly contains numerous edge-side data centers.A large number of wireless mobile terminal devices with storage,computation and communication functions constitute the bottommost terminal layer.They form a number of edge networks with the edge-side data center in a self-organizing manner.2.Research on optimization technique of computation task offloading for dispersed computing architecture.In order to ensure the superior performance level of the computing system and ensure the stability of the entire computing system ecology,issues such as competition for system resources and energy constraints of wireless terminal equipments need to be resolved.Therefore,based on the dispersed computing architecture integrating storage,computation and communication,this paper studies the key technology of resource optimization – computation task offloading optimization,and uses game theory to propose a cloudlet-based computation offloading algorithm for limited resources.This paper also verifies the performance of the proposed algorithm in terms of energy consumption,scalability,time efficiency and dynamics through a large number of simulation experiments.3.Design,implementation and research of the dispersed computing architecture system platform.Based on the above research,this paper designs,implements and studies the system platform for the dispersed computing architecture that integrates storage,computation and communication.Specifically,this paper integrates the proposed offloading strategy and designs and implements key system modules to build a dispersed computing architecture simulation platform integrating storage,computation and communication.On this basis,this paper tests and evaluates the simulation system platform through a large number of experiments,analyzes and studies the performance of the proposed algorithm in terms of task failure rate and average network delay,providing the system platform verification for the dispersed computing architecture integrating storage,computation and communication.The computing architecture and its optimization technology are important research directions for vertical industry interconnection applications and even the future Internet of Everything scenarios.As people's lifestyles changing and upgrading,it is especially important to provide efficient infrastructure support for connected application services.Based on three aspects of the computing architecture,the key technology for resource optimization and the system platform verification,this paper studies the dispersed computing architecture integrating storage,computation and communication,proposes a multi-layer game computation task offloading strategy,and designs and implements a dispersed computing architecture system platform integrating storage,computation and communication.It is hoped that the work in this paper will bring new ideas to the next generation of computing and network architecture and contribute to the research and advancement of future computing architecture.