基于多元组的格式化数据存储模型

Multi-Tuple Based on Formatted Data Storage Model

大数据时代,数据体量庞大、数据价值密度低、数据种类多样,如何从海量数据中挖掘有用信息成为研究重点。大量数据信息中格式化数据仍然占据很大比例,主要有数据标准规范不一、各业务数据单独存储且互不关联等问题,给数据治理、关联融合、挖掘分析产生了极大的障碍,难以发挥数据价值。提出了一种基于多元组的格式化数据存储模型,该模型通过对原始格式化数据拆分实现数据格式的统一,通过引入业务属性信息实现了对原始数据的分类管理。应用示例表明,该模型可有效解决各业务数据单独存储、互不关联的问题,可扩展性强、设计实现简便,能够为数据治理、数据关联融合、挖掘分析提供有效支撑。

In the big data era,data is characterized by huge volume,low value density and diversity type.Mining useful information from massive data has become the focus of research.In large amount of data and information,formatted data still occupies a large proportion,which mainly faces the prob-lems of different data standards and specifications,separate storage,and no association with each other.This has created great obstacles to data governance,data association,mining and analysis,making it difficult to leverage the value of data.The paper proposes a formatted data storage model based on multivariate groups.The model realizes the unification of data format by splitting the origi-nal formatted data,and achieves the classification management of the original data by importing bus-iness attribute information.It effectively solves the problem that each business data is stored sepa-rately and not related to each other,with strong scalability,simple design and implementation.It can provide effective support for data governance,data association and fusion,mining and analysis.