Research And Application Of Distributed Storage System Based On Cloud Computing

The time of internet infrastructure construction has passed, internet is entering a new economic era and facing a new changing which can provide an increasing number of services. Mobile internet and internet combination make the coverage of internet more extensive. The generation of cloud computing make it possible to do a variety of applications just using one platform. At the same time, as the development of network bandwidth and other network technology, accessing the non-local services through network is becoming more mature which promote the development of cloud storage. The non-local services including data processing, storage and information services. The research on the system has a very high application value.The thesis is mainly to research cloud storage system which based on HDFS. It aims to solve enterprises' massive data storage problem, reduces cost of applying distributed file system and promotes the development of Hadoop technology. Hadoop is a collection of related sub-project about distributed infrastructure including Hive, HBase, Pig, and Chukwa. They provide ancillary and supplementary services. The core framework design is HDFS and MapReduce. Hadoop distributed file system (HDFS) is designed as a file system which is suitable for running on generic hardware. It provides underlying support for Hadoop distributed computing and storage.The main purpose of the thesis is to achieve a cloud storage service system which can solve problems like unstructured data online storage, query and backup. It proposes a solution to solve the signal NameNode failure of Hadoop cluster. In order to make sure the cluster running normally, the NameNode can switch form crash node to backup node automatically. This thesis has done a comprehensive and detailed research on Hadoop platform, but how to protect the data security and privacy, optimize the Hadoop cluster and program efficiently to processing the large data are still the problems which need to be solved urgently in the future.