Research On Cloud Data Placement Method For Cloud Storage System

Cloud storage is derived,extended and developed on the basis of cloud computing.It is a cloud computing system which centered on data storage and data management.It is widely used to store and manage massive data in the era of big data.Cloud storage is a storage pool consist of a large number of heterogeneous storage devices which connect through the network.It provide a flexible,transparent and on-demand storage resource allocation cloud system for authorized users through a unified Web service interface.In the face of the bad network environment and the actual needs of the users,there are some evaluation indicators to evaluate the performance of the cloud storage.Data storage security and data retrieval time are two important indicators of data storage in cloud environment.It is important that how to minimize data retrieval time on the premise of ensuring storage data security.From the point of view of target optimization,this is a multiobjective optimization problem(MOP).In order to solve the multi-objective problem of cloud data storage,this thesis proposes a secure cloud data placement method based on multi-objective genetic algorithm,this method defines a security level associated with the data block and the data node distance,and then construct a multi-objective constrained optimization model to maximize safety level and minimize the retrieval time,the model considers data nodes,network bandwidth and shortest path etc..Finally,the Genetic Algorithm(GA)is adopted to solve this constrained multi-objective problem,so as to get the optimal cloud data security placement strategy.At last,a simulation experiment to this strategy has been carried out,the experimental results show that security cloud data placement strategy which in this thesis has a well balance on the data security and retrieval efficiency.It can minimize the data retrieval time and meet the demand of data storage security.