Multi-service correlation analysis method and device based on metadata graph structural similarity

Abstract

The invention discloses a multi-service correlation analysis method and device based on the structural similarity of a metadata graph. The method includes the steps of after obtaining metadata from multiple services, building a relation graph of a metadata object; judging whether or not a common metadata object and a common metadata object property exist in the same meta-model in the relation graph of the metadata object; if yes, obtaining the structural similarity of the metadata graph according to structural vertexes, the properties of the vertexes and similarities of edges in the relation graph of the metadata object. Based on the structural similarity of the metadata graph, the correlation relationships among the multiple services are confirmed.

Description

technical field [0001] The invention relates to a correlation analysis technology, in particular to a multi-service correlation analysis method and device based on the structural similarity of metadata graphs. Background technique [0002] Metadata refers to data that describes data, mainly describing concepts, relationships, rules, semantics, etc. of a domain. Metadata is an effective way to manage massive data systems (such as data warehouses, data marts, Hadoop big data platforms, etc.), it can provide a clear and complete catalog for accessing data, so that users can clearly understand the data as a whole and guide users Use data efficiently. [0003] When using the existing technology to conduct correlation analysis based on metadata, there are mainly the following defects: [0004] First, a relational link of metadata is a business process that has a direct reference relationship or a direct data flow relationship from the beginning to the end. In general, there are ...

AI Technical Summary

Problems solved by technology

[0004] First, a relational link of metadata is a business process that has a direct reference relationship or a direct data flow relationship from the beginning to the end. In general, there are many indirect links between multiple businesses of an enterprise, but the existing The metadata system has no way to determine the key connection points between multiple businesses, so when a business caliber changes, it is impossible to intuitively evaluate the impact of this business on other businesses, and can only use the manual backtracking method to find each metadata object Impact on business processes

[0005] Second, the existing metadata correlation analysis only roughly compares the number of overlapping metadata objects in two relationship links. In fact, different businesses often use different attributes of metadata objects and the logical flow relationship is often different. , therefore, the results of association analysis that do not consider the attribute information and logical relationship of metadata objects often lack accuracy

[0006] 3. At present, the classification of business mainly depends on the manual classification of business personnel based on their experience. When the amount of data is small, it can be handled reluctantly. However, in the face of massive and complex big data, manual classification is obviously powerless, and the existing metadata application system Lack of methods and mechanisms to assist classification

Images