海量异构科技文献信息资源的非结构化存储研究

RESEARCH ON UNSTRUCTURED STORAGE OF MASSIVE HETEROGENEOUS SCIENTIFIC AND TECHNOLOGICAL LITERATURE INFORMATION RESOURCES

在大数据环境下,科技文献信息资源具有分散、异构、动态、海量等特性,为解决传统关系型数据库存储这些数据时面临的不易扩展、性能较低等难题,进行了以非关系型数据库为核心的异构科技文献信息资源存储研究。形式化描述科技文献资源的结构;在此基础上研究非关系型数据库存储异构文献信息资源的可行性数据模式;结合科技文献信息资源的特点设计面向异构资源的No SQL(Not Only SQL)适配器来融合现有存量信息资源。实验验证了No SQL适配器的高效性以及非关系型数据库存储科技文献信息资源时的可行性、高性能以及高扩展性。

Under the big data environment, scientific and technological literature information resources are characterized by decentralized,heterogeneous,dynamic,and mass. In order to solve the difficult problems such as difficult to expand and lower performance when the traditional relational database stores these data,the research on the storage of heterogeneous scientific and technological literature information resources centered on non-relational databases had been carried out. First,we formally described the structure of scientific literature resources. And based on this,we studied the feasibility data model of non-relational databases storing heterogeneous literature information resources.Then,the No SQL adapter for heterogeneous resources was designed to fuse existing inventory information resources with the characteristics of the scientific literature information resources. Finally,experiments verified the high efficiency of the No SQL adapter and the feasibility,high performance,and high scalability of non-relational database storage of scientific and technical information resources.