TECHNIQUES FOR CORRELATING DATA IN A REPOSITORY SYSTEM

摘要

Techniques are described for determining correlations between data in a repository system. The data may include information about corresponding to resources (e.g., an application, a process, a service, an endpoint, or a method) in a computing environment. A correlation between objects can indicate a similarity or a relationship based on one or more of the attributes of each object that is correlated. The repository system can store information about each object in a data structure, such as an entity, including the attributes about the object. The repository system can determine the relationships between entities based on correlations identified from the attributes of entities. The repository system can perform correlations based on groups of entities corresponding to a group of objects. Upon determining that two groups of entities match, the repository system can compare individual entities in the groups to identify correlations between individual entities corresponding to objects that are correlated.

主权项

1. A method comprising:
identifying, by a computing system, a group of source entities from a plurality of entities in a source repository system; accessing, by the computing system, a group of destination entities in a destination repository system; determining, by the computing system, whether the group of source entities has a correlation with the group of destination entities; and upon determining that the group of source entities has a correlation with the group of destination entities, for each source entity in the group of source entities:
updating, by the computing system, the destination repository system to include the source entity in the group of destination entities when the source entity is determined to be a duplicate of a destination entity in the group of destination entities.