企业数据空间的数据组织方法研究-《计算机技术与发展》

文章信息/Info

Author(s):: WEN Bi-long; JIAO Sheng-jie; GUO Jiao; School of Computer and Information Technology,Northeast Petroleum University,Daqing 163318,China

Keywords:: enterprise dataspace; data organization; data resource catalog; property graph; data model

摘要:: 企业数据空间的主体是整个企业,面向多个部门、专业或项目,数据规模巨大、种类复杂多样,还需要管理大量已有严格数据模式的数据库,数据组织管理困难。然而现有数据空间的数据组织方法,大多是基于个人数据空间的,无法满足企业数据空间复杂的数据管理需求。为了统一描述多源异构数据、多维多角度地灵活组织数据和将传统的“先模式后数据”和数据空间的“先数据后模式”方式协调起来进行管理,提出了企业数据空间的数据组织方法:通过构建分层组织模型实现对数据多维多角度地灵活组织;利用属性图模型对企业数据空间中的各种数据资源进行统一描述和管理。基于该方法可以更好地描述企业中的各种数据资源,为企业提供灵活高效的数据组织和管理方式,进而更好地支持企业数据空间的数据模式演化,提高企业的数据管理效率,满足企业的数据组织管理需求。

Abstract:: The main body of the enterprise data space is the entire enterprise,facing multiple departments,professions,or projects. The scale of the data is huge,and the types are complex and diverse. It also needs to manage a large number of databases with strict data models,which makes data organization and management difficult. However,the existing data organization methods of data space are mostly based on personal data space, which cannot meet the complex data management requirements of enterprise data space. In order to uniformly describe multi-source heterogeneous data, organize data flexibly in multiple dimensions and multiple angles,and coordinate the traditional “first model first data” and data space “first data first mode” methods for management,a data organization for enterprise data space is proposed. A multi-dimensional and multi-angle flexible organization of data is achieved by constructing a hierarchical organization model; a unified description and manage-ment of various data resources in the enterprise data space is made by using the attribute graph model. Based on this method,various data resources in the enterprise can be better described, and flexible and efficient data organization and management methods can be provided for enterprises, so as to better support the data model evolution of the enterprise data space,improve the data management efficiency of the enterprise,and meet the needs of enterprise data organization and management.