开放可伸缩关系数据模型

Open Scalable Relational Data Model

数据模型研究是大数据领域中的一项基础性工作,能够为上层数据库构建、数据存取、数据分析和挖掘提供有力支撑.首先,针对传统关系数据模型数据类型受限,大数据集查询性能低和横向伸缩性不足的缺陷,提出一个开放可伸缩关系数据模型,该模型保留并扩展了传统关系模型的关系描述能力,提供开放的数据类型支持,并借鉴key-value的思想,提供了完全的横向伸缩特性,迎合了大数据体量庞大、类型多样、增长迅速的特点.然后,从功能特性和性能指标两方面,将OSRDM与主流数据模型进行分析对比,评价结果说明OSRDM的优越性.

In the field of data science, data model research plays a fundamental role that strongly supports the upper levels of database construction, data access, data analysis and mining. However, in traditional databases, there are restrictions of limited data types, low query performance and lack of horizontal scalability. Considering these factors,this paper puts forward an Open Scalable Relational Data Model. This model brings an open support for data types and complete horizontal scalability on basis of keeping and extending the relation descriptive power of the traditional relational data model. This caters to the characteristics of big data such as volume, variety and velocity. Then OSRDM is analyzed and evaluated by being compared with popular data models from two aspects：functional features and query performance. The result shows the superiority of OSRDM.