Cloud Replica Strategy Based On Multi-objective Optimization

With the rapid development of Internet technology and the arrival of 5G era,cloud computing and cloud storage are gradually integrated into our lives.Due to the limited personal storage resources and the convenience of cloud storage resources,many users choose to purchase services from cloud storage service providers.The cloud copy strategy directly affects the reliability and quality of service of cloud storage system.Therefore,the research of cloud copy strategy is one of the hot spots of cloud storage technology.Setting a reasonable number of copies and choosing the right location for the copies is an effective way to improve the quality of service.Through the research on the existing cloud copy strategy,the research results of this thesis mainly include:(1)To set the number of replicas more reasonably,this thesis proposes a cloud copy number determination algorithm based on IOWA operator.In this algorithm,the combined prediction model based on Iowa operator can predict the next periodic access of the file according to the historical access of the file,and get the next periodic access of the file more accurately.In order to calculate the number of increase or deletion of the next cycle,the number of copies required for the next cycle of the file is calculated using the predicted number of accesses to the next cycle of the file,and the number of copies of the file in the current cycle is considered.According to the experimental data,the correctness and effectiveness of the algorithm are verified.(2)In order to improve the quality of service of cloud storage system,this thesis consider the impact of the location of the replica on the access delay,load balancing and reliability of the storage system.In this thesis,a multi-objectives optimal replica placement strategy is proposed to reduce the access delay,reduce the load difference of storage nodes and improve the system reliability.The strategy obtains the approximate optimal solution of replica layout by solving the multi-objectives optimization function with the cuckoo algorithm.Finally,this thesis designs and implements the cloud copy visualization system through the SSM framework,validates the algorithm proposed in this thesis,and presents the data information stored in the cloud in a visual way.