Mass data processing system based on large-scale low-cost computing platform
基于大规模廉价计算平台的海量数据处理系统的研究*

This paper proposed a new mass data processing model based on large-scale low-cost computing platform, drawing on the basic idea of Map/Reduce the distributed computing model and Bigtable the large-scale data storage mechanism to achieve a data-center and computing-intensive economy supercomputing system platform. The system chose the business data of telecommunications as its analysis object, processing the large data sets of voice and data services. The system is suitable for processing other kinds of mass data, which provides valuable analysis services for telecom operators and ordinary users and offers a good reference model to other applications.