Distributed storage based on cloud computing

Cloud computing is the next generation of computing model, which plays an important role in the scientific computing and commercial field, and gets attention from the academic and business circles. Distributed storage of data in cloud computing environment in data center of the organization and management, is the core infrastructure of cloud computing environment. In addition, data centers typically consists of millions of nodes, and the scale of the data stored on it often reach PB level or EB, resulting in a normal behavior of data failure data, which greatly limits the application and popularization of cloud computing. Therefore, to improve the scalability fault tolerance to reduce cost, some key technologies of storage distribution of a cloud computing environment are come up with, in order to improve the storage scalability, fault tolerance and reduce memory energy consumption targets, First, this paper introduced and compared the advantages and disadvantages of data center network structure type. Secondly, it introduced and compared the current commonly used two kinds of distributed storage fault-tolerant techniques. Finally, it pointed out the current technologies, the major challenges and the direction of next research.