The Research For Data Storage And File Management Of Cloud Computing Platform

With the rapidly development of digital technology and Internet, the Internet could provide more and more types of services, and also should deal with amount of data. How to handle the vast amounts of data and services, and provide convenient and efficient network services for users become increasingly important. The arrival of cloud computing to solve the problem of massive data provide a good way, it provides a secure and reliable data storage, fast and convenient Internet services and power computing, it also has changed the traditional PC-based production mode, and ultimately change the way people get information. Meanwhile, with the promotion and popularity of cloud computing, how high-speed, low-cost storage and management of large amount of data that generated in the cloud, have become a major research focus of enterprises and organizations, which requires a good design of architecture and cloud storage platform.The current storage design of cloud computing platform mainly include Google' GFS and Hadoop'HDFS. However, there is the problem of node failure in the GFS and HDFS. Node failure will have an adverse impact on the cloud platform, especially the server node failure would course greater impact. So to solve single-node or dual nodes become an important practical issue.Firstly, this paper introduces the basic concepts of cloud computing and a number of technologies such as virtualization and P2P technology, introduces the current cloud computing platform and storage.then, for the node failure problem, this paper uses a combination of cloud computing and P2P architecture platform, that using cloud computing services store between nodes and the children servers in the data store, among the server nodes, a fully distributed between the nodes makes the service node type unstructured full reciprocity between the Server node model, which is a hybrid topology mode. The unstructured system using a fully distributed topology, for redundancy and delay considerations, most of the DHT always copy backup redundant information on the node that the virtual identity and the keyword are the nearest, this also avoids the problem of single node failure. This hybrid topology model can effectively solve the problem of the server node failure, and build trusted server node cloud computing infrastructure.Meanwhile, as it is unprecedented for cloud computing center to face the application types, numbers and scale, so the application and the user isolation technology is the cloud computing center need to consider. The Data Isolation in isolation technology is the most important, we use a hierarchical user isolation method based one-way pointer of directory mapping in order to facilitate effective implementation of the user data isolation.Then, this paper focuses on the analysis of data storage way of cloud computing platform, which is different from relational database and is suitable for cloud computing mapping data storage based on listed models that data table structure is a sparse, distribution, continuously multidimensional sorting mapping array.Finally, it provides the system architecture of cloud storage platform design, distribute storage process and implementation method, module design of system management nodes, child nodes and client nodes.