Research On High Reliability Methods Based On Network Coding For Cloud Data Storage

With the development of cloud computing and big data,cloud computing service is increasingly applied.As one of the most important parts of cloud computing,cloud storage has brought great convenience to data storage and service providing.However,with cloud computing services into the life of people,cloud storage is also facing many challenges,especially in recent years a number of data security incidents,how to ensure the safety and reliability of the data stored in cloud has become an urgent problem.Therefore,a security cloud storage framework with high recoverability based on network coding is proposed in this paper.Base on this framework,a secure cloud storage method with dynamic self-adapting based on nodes reliability and a data recovery method for multi node based on three-dimensional model are proposed,which are used to improve the reliability of cloud data storage,especially to solve the problem of multi nodes failure.Firstly,a security cloud storage framework with high recoverability based on network coding is presented,which can encode divided data with network coding,reduce the redundancy of storage space by analyzing the historical information of storage nodes and storing dynamically based on reliability value of nodes.The framework presented in this paper contains a three-dimensional model,which is utilized to improve the integrity of data and recover damaged data effectively.Secondly,in order to avoid storing data in the nodes of low reliability,a secure cloud storage method with dynamic self-adapting based on nodes reliability is presented.The method allocates data blocks dynamically by evaluating the reliability of storage nodes,which can also back up the data of nodes with low reliability on other storage nodes.The experimental results show that the method can reduce the storage space which doesn't need to back up all the data,and distribute data blocks more proportionate,then ensure the safety of data stored.Finally,due to the present cloud storage methods usually solve data recovery for single node failure,a data recovery method for multi nodes based on three-dimensional model is proposed.The method distributes data on the three-dimensional model that makes data associated with each side of the model.When many nodes are failure,the method can recover data efficiently.From experimental results,comparing with traditional copy storage methods.The method can solve data recovery when multi nodes failure,which can improve the recovery rate of cloud data and have higher reliability.