Browse Prior Art Database

A technique for importing metadata with duplicates and synching up from external sources into metadata server

IP.com Disclosure Number: IPCOM000236193D
Publication Date: 2014-Apr-11
Document File: 4 page(s) / 43K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is an automated mechanism for editing the identity of metadata assets with the same identity while importing the metadata from external sources into a metadata repository to make them unique in the metadata repository, and also define an automated mechanism for reconciliation of metadata assets with the metadata assets in the repository when the metadata assets have same identity.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 38% of the total text.

Page 01 of 4

A technique for importing metadata with duplicates and synching up from external sources into metadata server

Disclosed is a technique to detect and cleanup the duplicates automatically at the time of importing into the metadata repository and automatic reconciliation of these duplicates when they are re-imported.

Motivation:


Many sources of metadata allows duplicate names for the metadata assets.

- Erwin modeling tool allows 2 indexes created for a table with the same name
- Erwin modeling tool allows data values in a data item definition without names,

just values.

- Embarcadero ER/Studio Data Architect allows 2 foreign keys created for a table with the same name

The tools that use metadata fail when there are multiple assets with the same business identity.

- In most cases, the business identity of an asset is the name of the asset and the business identity of the container. For example, the identity of an index is the name of the index and the identity of the table that contains the table and the identity of the table is the name of the table and the identity of the schema that contains the table etc.

It is not practical to manually resolve the duplicate names in the metadata repository.
- the normal size of the metadata assets can be as big as 50,000.
- the user need to perform this manual step every time the metadata is re-imported from external sources.

It is not acceptable for expecting the user to resolve the duplicate identities in the external sources before importing metadata into metadata server.
- the user importing the metadata may not be the author of the model.
- the normal size of the metadata assets can be as big as 50,000.

Reconciliation of assets with duplicate identities is not possible when the metadata is re-imported from external sources.
- if the metadata repository has 2 foreign keys with the same name and a foreign key with the same name is re-imported, then it is not possible to decide which foreign key in the repository should be merged with the imported foreign key.

Reconciliation is not possible when the metadata with duplicate identities is transferred from one metadata repository to another metadata repository, for the same reason given in the previous point.

Problems with the existing solutions:

1


Page 02 of 4

Existing Solution in the market does not provide mechanism to identify and clean the metadata being imported into metadata server when the name of the objects are same or missing

Leads to problem :
- metadata objects are saved as duplicates
- tools that use the metadata fail because of these duplicates

Problem Definition:

Need an automated mechanism for resolving metadata assets with the same identity while importing the metadata from external source into a metadata server, so that the metadata asset can be uniquely identified by its identity.

In this solution, we propose a technique to resolve the identities while importing metadata with duplicate identities from external sources, so all the asse...