2022Unpublished venueRequires access

M-Data-Fabric: A Data Fabric System Based on Metadata

Kang Liu, Mingchuan Yang, Xinchi Li, Kaicheng Zhang, Xiaoqing Xia, Hui Yan

Open publisher page 14 citations

Abstract

The data has become an island within the application, indicating that the data has not been fully utilized. Throughout the current performance ranking, almost all enterprises are data-driven enterprises. However, several challenges prevent them from making full use of all data, lack of data access, numerous data sources and data types, and complex data integration. Research shows that in most enterprises, up to 74% of the data has not been analyzed, and up to 82 % of enterprises are restricted by data islands. In order to solve the above problems, this paper designs a metadata based data fabric system architecture (M-Data-Fabric). M-Data-Fabric realizes the overall use of different data sources rather than reuse according to the process of metadata identification, analysis, integration, integration and knowledge map construction. By analyzing the relationship between data sources, metadata is aggregated through strong business After correlation, the whole business knowledge map is obtained.

About this research paper

What this paper is about

The data has become an island within the application, indicating that the data has not been fully utilized. Throughout the current performance ranking, almost all enterprises are data-driven enterprises. However, several challenges prevent them from making full use of all data, lack of data access, numerous data sources and data types, and complex data integration. Research shows that in most enterprises, up to 74% of the data has not been analyzed, and up to 82 % of enterprises are restricted by data islands. In order to solve the above problems, this paper designs a metadata based data fabric system architecture (M-Data-Fabric). M-Data-Fabric realizes the overall use of different data sources rather than reuse according to the process of metadata identification, analysis, integration, integration and knowledge map construction. By analyzing the relationship between data sources, metadata is aggregated through strong business After correlation, the whole business knowledge map is obtained.

Why it matters

OpenAlex reports 14 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The data has become an island within the application, indicating that the data has not been fully utilized. Throughout the current performance ranking, almost all enterprises are data-driven enterprises. However, several challenges prevent them from making full use of all data, lack of data access, numerous data sources and data types, and complex data integration. Research shows that in most enterprises, up to 74% of the data has not been analyzed, and up to 82 % of enterprises are restricted by data islands. In order to solve the above problems, this paper designs a metadata based data fabric system architecture (M-Data-Fabric). M-Data-Fabric realizes the overall use of different data sources rather than reuse according to the process of metadata identification, analysis, integration, integration and knowledge map construction. By analyzing the relationship between data sources, metadata is aggregated through strong business After correlation, the whole business knowledge map is obtained.

Key concepts: Metadata, Data element, Data mapping, Metadata repository, Computer science, Data integration, Metadata management, Data architecture

Related papers

Back to paper searchBrowse research topicsOriginal source
M-Data-Fabric: A Data Fabric System Based on Metadata — Research Paper | ScholarLens