Introduction

Getting agreement across an organization on the definition of terms such as Net Sales and Customer can be a challenge. An even bigger challenge is developing a common understanding across business units and across business and IT. Yet such an understanding is critical to effective warehouse and report design, to master data management initiatives, and to consolidation resulting from M&A activity.

Business metadata documents the business meaning and categorization of data assets. It is defined in business language, and it is independent of technology. InfoSphere Foundation Tools let you establish responsibility, accountability, and traceability for each business term or category and enable collaboration among business analysts, business users, data modelers, and other IT professionals.

InfoSphere Foundation Tools include:

• IBM InfoSphere Information Analyzer
• IBM InfoSphere Data Architect
• IBM InfoSphere FastTrack
• IBM InfoSphere Metadata Workbench
• IBM InfoSphere Business Glossary, including Business Glossary Anywhere

The Unified Metadata Management layer within InfoSphere is core-infrastructure designed for active metadata sharing across InfoSphere products. In Figure 1, Common Metadata, located in the center of the image, represents such a layer.

The InfoSphere Foundation Tools support three primary types of metadata:

• Business metadata is critical to providing context on an integration project. It's about defining terms in everyday language regardless of technical implementation. Business metadata typically includes definitions, terms, abbreviations, glossaries, classification, categories, examples, stewards, and owners, which are described using the business language. For example: "What is a customer order? How do I categorize a customer order?" The audience for this type of metadata is business users.
  
  
• Technical metadata is often used by more technical staff, such as developers. Technical metadata includes host server, database type, database schemas, table name, column names, and data types, all of which are described in technical detail. These objects are used heavily during the application design process. The audience for this type of metadata is specific tool users, such as for business intelligence, ETL, profiling, and modeling.
  
  
• Operational metadata refers to the metadata generated and captured when a process executes. Operational metadata is what enables the administrators to manage the system, troubleshoot issues, and ensure that things are running smoothly. It includes job name, job execution times, number of rows processed, error or success status, time started, and time completed, which are all described in sequence. The audience for this type of metadata is operations, management, and business users.

Each of these metadata types can be linked together to create an end-to-end relationship between them so that users can understand not just where information is stored, but they can also understand the business context of that information.

Information Server approaches the management of these three metadata types through a common architecture approach to storing, enhancing, and exchanging metadata, as shown in Figure 2.

• Store means that the unified metadata foundation of the InfoSphere Information Server platform which provides a single active repository to facilitate shared understanding across business and technical domains for metadata-driven integration accessed using a common services layer.
  
  
• Enhance means that the metadata stored within the common repository can be enhanced by creating links to third party metadata as well as implement controlled metadata extensions to capture additional business requirements.
  
  
• Exchange refers to the fact that InfoSphere Information Server is built on the concept of task-driven modules. Relevant metadata is exposed to and shared with each of these user roles using the appropriate interface.

The unified metadata management layer supports the following goals:

• All modules use the same active repository to share and exchange metadata effortlessly across user roles.
  
  
• Relevant metadata is shared and exchanged across user roles.
  
  
• Business context is added to technical metadata.
  
  
• Improved management and re-use occurs across integration projects.
  
  
• The integration process eliminates manual steps and maintenance of metadata.
  
  
• Organization confidence increases by providing insight into where information came from and what it means.
  
  
• Organizations can meet and exceed governance requirements, including compliance and regulatory mandates.

This metadata can be effectively shared and reused within Rational, InfoSphere, or Cognos® development projects. Next, this article describes how you can use some of the InfoSphere Foundation Tools to create, manage, and share key business metadata.

Inserting business metadata into InfoSphere Business Glossary

Business metadata can be created or loaded into InfoSphere Business Glossary. In the examples in this article, Information Server is installed, but it is not a prerequisite for using InfoSphere Business Glossary. If you have Information Server installed, you can access the InfoSphere Business Glossary using the Glossary tab inside the Information Server Web console. To access the Glossary tab, you need to log on through the Information Server Web console first. After you log in, you can see the screen shown in Figure 3.

To access the business metadata, click on the Glossary tab, as shown in Figure 3.

On the Glossary tab, you can browse, edit, import, and export business terms, categories, stewards, and custom attributes, as shown in Figure 4.

• Categories are used to organize business terms and other categories into hierarchies.
  
  
• Business terms define a common vocabulary between business and technical users. These terms, once defined, reviewed, and published, become the enterprise-standard vocabulary. You can include descriptions, examples, abbreviations, stewardship, synonyms, and related terms.
  
  
• Stewards are people who are responsible for documenting and maintaining the business terms and categories assigned to them.
  
  
• Custom attributes are used to customize the application to the users' needs.

In the Glossary, there are two options for creating business terms and categories: manually or by importing from a file (CSV or XML). Aside from these two options, you can also export glossary terms along with categories from glossary and physical models created in InfoSphere Data Architect using the export feature.

Creating categories manually

To create new categories, click Categories under Browse and Edit on the navigation frame, as shown in Figure 5.

The Categories view opens. In this view, you can browse, search, edit, delete, and create new categories or subcategories. Click New to create a new category, as shown in Figure 6.

After you click New, you can define the new category you want to create, as shown in Figure 7.

Provide the requested information:

• Name: Enter the name of the category. For example, you can use General terms. Categories are designed to be subject area folders that help end users find information easily. For multiple languages, each country or language should be a top-level category.
  
  
• Parent Category: Use this only if the category is a subcategory. Click Select, and choose the parent category. In the example, it is left blank. Lower-level categories should be organized according to business areas and according to how end-users would think about where to find the content. One example of a main category might be customer, and subcategories for customer might be high-value and low-value.
  
  
• Steward: Click on the icon, and select a user ID to assign a steward. In the example, it is left blank.
  
  
• Short Description: Enter text that explains the category. The example shows General terms used over the enterprise.
  
  
• Long Description: Enter a long description to explain the category in more detail. In the example, it is left blank.

During category creation, you can set custom attributes, contained terms, and other options by using custom attributes. You can define enterprise-specific properties for Categories and Terms. Some examples of Custom Attributes might be attributes such as HQ Approval Required, Basel II Term, and Link to Legacy Data Dictionary. To set this information, click on the links on the left of the screen, as shown in Figure 8.

After you complete the category information, click Save and Close. Figure 9 shows your recently created category (General terms) in the list.

Creating business terms manually

To create a new term, click the Terms link located in the Navigation frame, as shown in Figure 10.

The Terms view appears. Click New to add a new term, as shown in Figure 11.

Define the new business term. Most of the information is optional, but best practices encourage you to capture as much information as you can to fully document business terms. In the example, the term State Code is added, as shown in Figures 12 and 13.

After you complete the business term information, click Save and Close.

The new term is added to the Terms list, as shown in Figure 14.

Importing business terms and categories from a CSV file

If you already have the business terms and categories created in another document or software outside the InfoSphere Foundation Tools, export them to a CSV file. This file should be in the format shown in Listing 1. For the exercises in this article, you can use the sample CSV file (see Downloads).

 +++ Import Categories - Begin +++,,,,,,,
Category Name,Parent Category,Steward,Short Description,Long Description,,,
Customer,,,Captures information about customers,,,, +++ Import Categories - End
+++,,,,,,, +++ Import Terms - Begin +++,,,,,,, Term Name,Parent
Category,Steward,Short Description,Long Description,Usage,Example,Status
Gender,Customer,,Gender,,,,Accepted +++ Import Terms - End +++,,,,,,, 

The sample CSV file has two main sections: Categories and Terms. The Category section has five columns:

• Category Name
• Parent Category
• Steward
• Short Description
• Long Description

The Terms section has 12 columns:

• Term Name
• Parent Category
• Steward
• Short Description
• Long Description
• Usage
• Example
• Status
• Abbreviation 1
• Abbreviation 2
• Is Modifier
• Type

Here is a description of the code in Listing 1.

• +++ Import Categories - Begin +++,,,,,,,
  This first line indicates the beginning of the Categories sections.
  
  
• Category Name,Parent Category,Steward,Short Description,Long Description,,,
  The second line indicates the columns names to be used in import process.
  
  
• Customer,,,Captures information about customers,,,,
  The real data to be imported comes in the third row of each section, in this case the Categories section. Each column's data is separated from the next by a comma.
  
  
• +++ Import Categories - End +++,,,,,,,
  The last line in this section indicates the end of the Category section.
  
  
• +++ Import Terms - Begin +++,,,,,,,
  This line indicates the beginning of the Terms section.
  
  
• Term Name,Parent Category,Steward,Short Description,Long Description,Usage,Example,Status
  The second line indicates the columns' names to be used in the import process.
  
  
• Gender,Customer,,Gender,,,,Accepted
  The real data to be imported comes in the third row of each section. Each column's data is separated from the next by a comma.
  
  
• +++ Import Terms - End +++,,,,,,,
  The last line in this section indicates the end of the Terms section.

The import process for business terms and categories is simple and quick. Click the Import Terms and Categories link located on the navigation frame, as shown in Figure 15.

In the Import Terms and Categories view, provide the file name and path for the file containing the terms and categories, as shown in Figure 16.

Click Import to start the process. Figure 17 shows an example of the resulting output.

When you return to the Glossary tab home, you notice new categories and business terms. Once they are imported, you can edit them, and you can let the technical team members know about the new business terms.

Back to top

Importing metadata from Metadata Server into InfoSphere Data Architect

InfoSphere Data Architect can import and export metadata from and to the Metadata Server. This feature enables the enterprise to export an industry data model to the metadata server. A data modeler can import business terms and use them to associate these business concepts with conceptual and physical data model elements. To import the metadata, open InfoSphere Data Architect, and click File > Import, as shown in Figure 18.

Select Import a Glossary Model from the Metadata Server under the Data folder, and click Next, as shown in Figure 19.

Next, specify a project where you want to create the glossary model. Define the name for the new model, or specify which glossary model you want to update. Click Next. In the example, a new model is created called Glossary Model.ndm.

The Glossary Transformation Options screen appears, as shown in Figure 21. You can optionally transform this new glossary model into a logical data model, which can later be transformed to a physical data model by selecting the appropriate checkbox.

InfoSphere Data Architect needs more information before connecting to the metadata server, such as hostname, port number, user ID, and password, as shown in Figure 22.

Enter the requested information, and click OK.

Figure 23 shows the resulting Metabroker status window.

If the connection succeeds, the metabroker Metadata Prefilter window appears, as shown in Figure 24.

In the metabroker Metadata Prefilter window, you can filter which categories you want to import from the metadata server. Select which categories you want to import, and click OK.

The metabroker Status window appears again, as show in Figure 25. If successful, click Finish.

The new glossary model appears in the InfoSphere Data Architect, as shown in Figure 26.

Back to top

Exporting metadata from InfoSphere Data Architect into Metadata Server

The process of exporting metadata from InfoSphere Data Architect to Metadata Server is almost the same as importing data from Metadata Server into InfoSphere Data Architect. From InfoSphere Data Architect, you can either export a glossary model or a physical model to Metadata Server.

To show how to export a glossary model, Figure 27 shows that InfoSphere Data Architect has a Glossary Model opened.

To export the glossary to the metadata server and to make the glossary available as an InfoSphere Business Glossary, click File > Export. The Export window opens, as shown in Figure 28.

You have two export options under the Data folder:

• Export a Glossary Model to the Metadata Server: Export metadata from a glossary model to the metadata repository, where the metadata appears as categories and terms. In InfoSphere Business Glossary, use those terms to classify metadata objects in the repository.
  
  
• Export a Physical Model to the Metadata Server: Export metadata from a physical model to the metadata repository, where the metadata can be used by tools that interact with the repository. For example, you can use the imported table definitions in a job in IBM InfoSphere DataStage and QualityStage Designer, analyze the columns in IBM InfoSphere Information Analyzer, or assign stewards to the metadata objects by using InfoSphere Business Glossary.

For this example, select the Export a Glossary Model to the Metadata Server option. Click Next. The Export Model to Metadata Server screen appears, as shown in Figure 29.

In the Export Model window, choose from which Glossary Model the information should be exported. After you choose a model such as the Corporate Glossary Model, define a log file location, and click Finish. The metabroker screen appears, as show in Figure 30.

The Metabroker status window provides you with the options to export the entire Glossary Model into the Metadata Server, or you can filter what you want to export. For the example, export the entire glossary, because it is small.

After you define what will be exported, you can provide access information on the next screen, as shown in Figure 31.

After you provide the information requested, Data Architect connects to Metadata Server and exports the information you chose. The Metabroker status window appears again, letting you know about the success of this process, as shown in Figure 32.

If you want to confirm success, you can access the Glossary Tab in the Information Server. The terms and categories you exported should be there. In the example, the category is highlighted, as shown in Figure 33.

Back to top

Conclusion

The Unified Metadata Management layer is core-infrastructure designed for active metadata sharing across InfoSphere products. The layer enables integration between the tools used by business analysts and the tools used by technical team members, such as InfoSphere Data Architect and Information Server. The data architect can use business metadata to create or update an existing glossary model or a logical model.

Back to top

Download

Information about download methods

Resources

Learn

• Get more information about IBM InfoSphere Data Architect.
  
  
• Refer to "Migrate ERwin Data Models to Rational Data Architect" (developerWorks, September 2007) to migrate existing data models created in CA ERwin Data Modeler to RDA. Explore some of RDA's features that data modelers and data architects use on a daily basis.
  
  
• Check out "Use Rational Data Architect to define and enforce data object naming standards" (developerWorks, January 2007) to examine the features of Rational Data Architect that enable you to define and implement object naming standards, and then explore a real-world example.
  
  
• Read "Using Rational Data Architect to import and export XML" (developerWorks, Nov 2006) for a step-by-step guide on how to create XML files from Rational Data Architect, and how to create data models from XML files.
  
  
• Learn more about Information Management at the developerWorks Information Management zone. Find technical documentation, how-to articles, education, downloads, product information, and more.
  
  
• Stay current with developerWorks technical events and webcasts.
  
  

Get products and technologies

• Build your next development project with the InfoSphere Data Architect download, available directly from developerWorks.
  
  

Discuss

• Participate in the IBM Metadata Server, Workbench, and Business Glossary forum and the Rational Data Architect (now InfoSphere) forum.
  
  
• Get the scoop at the Data Studio Team blog.
  
  
• Check out the developerWorks blogs and get involved in the developerWorks community.
  
  

About the author

Denis Vasconcelos is a Data Specialist for IBM Global Business Services in Brazil. He is a co-author of "Tivoli Management Services Warehouse and Reporting" and "IBM Information Analyzer & Data Quality Assessment" IBM Redbook publications. His areas of expertise include database administration, data modeling, heterogeneous database migration, and project management. Denis has a Bachelor's degree in computer science and a post-graduate degree in project management.

Comments (Undergoing maintenance)

Back to top

Trademarks  |  My developerWorks terms and conditions

Table of contents

• Introduction
• Inserting business metadata into InfoSphere Business Glossary
• Importing metadata from Metadata Server into InfoSphere Data Architect
• Exporting metadata from InfoSphere Data Architect into Metadata Server
• Conclusion
• Download
• Resources
• About the author
• Comments

Next steps from IBM

• Try: InfoSphere Data Architect
• Demo: Integrated Data Studio in a real-world scenario
• Webcast: Five things you can do today to cut costs using agile development
• Briefing: Jazz and Rational Team Concert
• Buy: InfoSphere Data Architect

My developerWorks community

Dig deeper into Information Management on developerWorks

DB2 pureScale

See how true transparent scaling architecture can be achieved on distributed platforms

Special offers