Random Linear Network Coding Based Cloud Storage System Research

Compared with traditional computer clusters, the cloud computing model IaaS (infrastructure as a service) has many advantages in terms of flexibility and load bal-ancing. Besides these, IaaS can run in low-performance machines, this makes IaaS has good scalability. As an open source implementation of IaaS, OpenStack can manage large scale of computer clusters with virtualization technology. We will introduce the process of the deployment of OpenStack clouding computing platform.HDFS is a distributed cloud storage system that provides high throughput access to application data and is suitable for applications that have large data sets. HDFS is designed to run on commodity hardware, so we should focus on the availability of data if we store data on it. HDFS achieve high fault tolerance by a way of redundancy, which will occupy too much disk storage. We will provide a data redundancy scheme based on random linear network coding(RLNC), which applies to distributed storage system. It dramatically reduce data redundancy without reducing data reliability. With RLNC, third-party software is difficult to recover the original data from data hijacked, as a result that data security is guaranteed.Cloud storage can provide good service such as file synchronization service to users. As a representative of distributed storage system, HDFS can also provide quality of service to users. However, application developed on HDFS is seldom. We developed a tool called VISG cloud synchronization, which can synchronize files between local file system and HDFS. Besides, it is easily operated and user can start to use it without knowledge of HDFS. We will describe the design process, the main algorithm applied by and the usage instructions of the software in detail.