Design And Implemtation Of Distributed Storage System For Employer Information Service Platform

With the rapid development of information technology, network has become an important carrier of information. People have a great desire to get comprehensive information via network. The employment information is now dispersed in the local government departments across the country. Due to the mutual isolation between departments and technical limits, it is very difficult for job seekers and recruiters to get the information they need in a short time, which cause some social problem. In order to promote the national employment and recruitment service quality, and make full use of these data, there is an urgent need to build a unified data storage systems to provide high efficient, stable data storage service.This thesis investigates and analyzes the related technology of distributed storage the relevant classical distributed system. Based on the characteristics of the employer information, the author designs and implements a distributed storage system for employment information. The whole system is divided into three subsystems. For permanent storage of data, the system uses the open source distributed storage system, Cassandra. For the hot data caching, the subsystem deploys many cache services in different regions to accelerate and reduce the access function of distributed storage system pressure. For the scheduling of requests, the subsystem not only manage the whole system status in a reliable way, but also parses, analyzes, rewrites and forwards the request to the destination of storage and cache subsystem.Compared with the traditional relational database solutions, the system has the advantage in system scalability and availability. The performance of reading is also much better when the hot data set is limited. The research is a valuable reference to storage system, which is rapidly growing and has rigid requirements on system availability.