[Document] a database engine for mass data processing

A database engine for large-scale data processing

Wang Yi, Liu Changcheng, Ma Jianqing

When data volumes rise from GB to terabytes or even petabytes, parallel databases with high performance can have high computational costs while ensuring scalability and fault tolerance. To solve this problem, a parallel database engine flexdb for large-scale data processing is designed. The parallel computing framework of map Reduce is used as the communication layer to dispatch and coordinate the computation and communication of the nodes in the cluster. The experimental results show that the system performance of FLEXDB is close to the parallel database and has good scalability and fault tolerance.

Keywords: mass data; scalability; fault tolerance; Map reduce framework; parallel database

[Download Address]http://bbs.chinacloud.cn/showtopic-13016.aspx