Continue in 2 seconds

Interaction of the Complete Meta Data Model

  • May 01 2004, 1:00am EDT

Author's Note: This column is adapted from the book Universal Meta Data Models by David Marco and Michael Jennings (John Wiley & Sons).

In Part 1 of this series, you were introduced to the complete meta data model and provided a description of the components of the entire universal meta model (UMM) environment. In this month's installment, we will look at the principal interaction points between the various model areas. An understanding of how each of the model components interrelates is required before reviewing the individual components in more detail.

The complete model (see Figure 1) shows a composite view of the universal meta data environment, providing information that is more navigational in nature. The model is a derived logical data model. The complete model illustrates UMM components at an entity-package level of abstraction. The model provides a total view of the entity concepts and relationships, which are more important from a knowledge perspective. Let us look further at the detail of the entity packages and their interactions with other components.

Figure 1: The Complete Meta Data Model

In the center of the model, Data Package and Structure provides the meta data foundation for the rest of the components. The Data Package and Structure entity package represents the many physical forms data may take in an organization. Enterprise applications move data to and from any of these data forms through the direct relationship with the Data Transformation and the indirect relationship with Transform Process entity packages. The movement of data is the central notion for the enterprise. This view for modeling data movement among various forms is handled through the different sets of data having a common base structure. A specific form of data transformation for XML types is illustrated through the subtype relationship between Data Transformation and the subtype XSLT. A specific form of a data package for XML Schema (DTD) types is illustrated through the subtype relationship between Data Package and Structure and the subtype XML Schema (DTD). XSLT and XML Schema (DTD) are highlighted in order to focus on these transformation and data package types. Additionally, a record of data movement activity is captured through the relationship between Transformation Audit and both Data Transformation and Transform Processes.

Next, we look at the relationship between real-world objects and data that is illustrated in the relationships between System and Environment, and Data Package and Structure. The relationship of these two entity packages allows one to perform business analysis from an implementation management outlook plus see the data lineage and analyze impact throughout the firm. How data is associated to an application, whether purchased or developed in house, is shown through the connection between Software and Data Package and Structure. This area also shows which data structure provider is being used. The specific case of how message queues are processed and controlled through application components is illustrated in the link from Data Package and Structure to Business Transaction to Software. The explicit use of messaging as a data package type for a business transaction is illustrated through the connection of XML Schema (DTD) and Business Transaction.

The abstraction of software and data structure users is illustrated in the relationships between System, Hardware, Environment, Software and Organization. The association of these entity packages allows technology and architecture to be described and tracked. For example, business applications and databases are associated with deployment environments (e.g., development, quality assurance, user acceptance testing quality control, production, stress testing and training). The detail of the association of the hardware to the network has been summarized because a network comprises many interconnected hardware components. Service level agreement-related entities have been summarized in the relationship between Organization and Hardware. A project can be accomplished and tracked for a system by a person in an organization. A person belongs to an organization. These associations are illustrated through the relationship of Project to System and Person as well as those of Organization to System and Person.

The relationship of the business to information is shown through the links between Data Steward and the Subject Area, Data Package and Structure, and Business Rule entity packages. The link between a Data Steward and Person is removed for simplicity in the model view. The business view of accessing information is represented using subject areas that provide the connection to the more detailed data perspective. Data stewards also define the business meaning of data, hierarchies, calculations and inter-data dependencies through this area of the model. The detailed processing of data through expression-based business rules is illustrated in the relationship between Business Rule and Data Transformation.

The final area of focus in the complete model is data quality measurement. This is illustrated in the model through the dual relationships between Data Package and Structure through Domain to Business Rule, and Data Package and Structure through Data Quality to Business Rule. If data domain violations are encountered during detailed transformation or data profiling, the resultant domain or business rule violation can be tracked for further analysis. Depending on how this area is implemented, there could be a different rule for each data element or a single rule for all single element violations. Domain value violations could be single values, a range specification or a pattern mismatch.

In the next installment of this series, we will take a closer look at the enterprise systems portion of the complete meta data model.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access