• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2016, Vol. 38 ›› Issue (01): 33-40.

• 论文 • Previous Articles     Next Articles

Data switching cost between NoSQL databases 

GUO Kun1,SONG Jie1,WANG Jieping2,ZHU Zhiliang1   

  1. (1.College of Software,Northeastern University,Shenyang 110169;2.China Electronics Standardization Institute,Beijing 100007,China)(1.College of Software,Northeastern University,Shenyang 110169;2.China Electronics Standardization Institute,Beijing 100007,China)
  • Received:2015-08-13 Revised:2015-10-21 Online:2016-01-25 Published:2016-01-25

Abstract:

NoSQL databases get more and more widely used in the IT industry due to their excellent performance in business processes in big data environments. However, data organizations differ from each other because of different data models in each NoSQL database,  which incurs impedance during data transfer when we perform data switching between NoSQL databases. Business data encapsulated in source databasesdata models may not be directly parsed by the target database, an additional model adapting operation therefore is needed. In this paper we attempt to define a data description model, which models the features of NoSQL databasesdata model, and describes the way NoSQL databases organize data, and then defines the corresponding distance evaluation algorithm. According to the data description model and the distance evaluation algorithm a unified data model can be designed and developed, which can transform to all the data models of NoSQL databases during data switching process, thus the related business code design just needs to align with the unified data model rather than each concrete data model.

 

Key words: NoSQL;data switching;data model;distance evaluation;edit graph