The traditional data analysis, it is difficult to meet data processing efficiency requirements at this stage. Application of Hadoop cloud computing technology, realization of data storage and analysis, improve the efficiency of data storage and analysis. This paper is on the basis of summing up the pros and cons of traditional system, with Hadoop Distributed File System (HDFS) to replace the existing stand-alone data storage, map/reduce application instead of the traditional stand-alone data analysis, and make the optimization. The experiment approves that the feasibility of Hadoop system architecture in the production deployment and using.