Browse Prior Art Database

Universal Metadata Repository

IP.com Disclosure Number: IPCOM000015116D
Original Publication Date: 2002-Jan-17
Included in the Prior Art Database: 2003-Jun-20
Document File: 4 page(s) / 105K

Publishing Venue

IBM

Abstract

UNIVERSAL METADATA REPOSITORY INTRODUCTION Disclosed is a generic but infinitely extensible database schema that has a data driven approach to creating and managing a metadata repository. Metadata is recognized as a key asset of the business and technical information environment however few organizations have been able to effectively and easily organize and manage it. Current metadata repositories are rigid, closed information stores that are data store and database schema specific to a vendor’s solution set requiring either a vendor specific approach to metadata management or specialized customization. The Universal Metadata Repository (UMR) addresses these short comings of past and current metadata stores or repositories by creating a generic but infinitely extensible database schema which has a data driven approach to creating and managing metadata. The UMR is based on an abstracted or generalized data model that is highly extensible and allowing the capturing of any type of metadata both common metadata such as Logical Data Model Entities and Attributes, Physical Database Tables and Columns and “newly discovered” Types of metadata without changing the underlying repository structure. This is done by adding rows/records to the repository to first define the metadata Type(s) generally and then adding rows/records about the specific instance information/data thus capturing these “new” Types of metadata.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 38% of the total text.

Page 1 of 4

Universal Metadata Repository

UNIVERSAL METADATA REPOSITORY INTRODUCTION

Disclosed is a generic but infinitely extensible database schema that has a data driven approach to creating and managing a metadata repository.

Metadata is recognized as a key asset of the business and technical information environment however few organizations have been able to effectively and easily organize and manage it. Current metadata repositories are rigid, closed information stores that are data store and database schema specific to a vendor's solution set requiring either a vendor specific approach to metadata management or specialized customization.

The Universal Metadata Repository (UMR) addresses these short comings of past and current metadata stores or repositories by creating a generic but infinitely extensible database schema which has a data driven approach to creating and managing metadata. The UMR is based on an abstracted or generalized data model that is highly extensible and allowing the capturing of any type of metadata both common metadata such as Logical Data Model Entities and Attributes, Physical Database Tables and Columns and "newly discovered" Types of metadata without changing the underlying repository structure. This is done by adding rows/records to the repository to first define the metadata Type(s) generally and then adding rows/records about the specific instance information/data thus capturing these "new" Types of metadata.

The result is the metadata modeling, application development, and maintenance efforts (time and associated costs) are greatly reduced and the limits imposed on the types of metadata stored constrained only by the needs of the business and that imposed by database internals and/or hardware capacities.

To better understand the URM this document describes release 1.0 of the UMR data model and articulates a sample implementation of the repository.

Universal Metadata Repository Data Model Overview

The foundation of the UMR Model lies in object theory leveraging the concept of objects and the relationship of these objects. Enterprise business and technical information is metadata. This metadata information or structure is represented in the UMR as objects. The information about the instances of these metadata objects can be fully captured by storing the aspects of this information (when applicable) as objects, the object properties, object types, object type properties, relationships, relationship properties, relationship types, relationship type properties, and primitive type information. These aspects of metadata information physically manifest as data model entities in a one-to-one relationship.

Data Model Entities

1

Page 2 of 4

Listed below are the entities making up the UMR data model and their definitions. Please reference the Logical Data Model diagram following this listing.

Object Type Entity

The entity Object Type captures a classification scheme for Objects with attributes for the object's name a...