Model management tasks

When you develop enterprise models, perform typical model management tasks, such as propagating, versioning, and harvesting to allow parallel development of projects and maintain your enterprise, projects and practitioner model files.

About this task

Note: this user guide only identifies the model management tasks that involve the IBM Industry Models when you develop the data models, and does not describe how to perform these tasks with a particular tool, such as Rational Team Concert.
Figure 1. Model management tasks examples
Model management tasks examples
This figure illustrates the streams (a stream is a logical group of files) that are used when working with the IBM Industry Models:
IBM Industry Models stream
A stream with the versions of the IBM Industry Models files.
Enterprise stream
A stream that descends from the IBM Industry Models stream, and includes the versions of your enterprise files.
Project streams
A stream that descends from the Enterprise stream, and includes the versions of the project files. There is one such stream per project.
Practitioner streams
A stream that descends from a project stream, and includes the versions of a practitioner files. There is one such stream per practitioner within a project.
The figure also illustrates one scenario of model management tasks that can take place when you develop a data model:
  1. Propagate the IBM Industry Models version 1.0 to the Enterprise stream.
  2. Propagate the enterprise version 1.0 to the Project 1 stream.
  3. Propagate the enterprise version 1.0 to the Project 2 stream.
  4. Project 1 customizes the data model and creates version 1.1.
    Note: the Practitioner stream is not illustrated in the figure, but is implied.
  5. Version 1.1 is harvested to the Enterprise stream without any changes since no changes happened in the Enterprise stream during Project 1 customization.
  6. Propagate the project version 1.0 to the Practitioner A stream.
  7. Practitioner A customizes the data model and creates version 1.2 and 1.3.
  8. Version 1.3 is harvested to the Project 2 stream without any changes since no changes happened in the Project stream during Practitioner A customization.
  9. Version 1.3 is harvested to the Enterprise stream. This request a merge with v1.3, the latest enterprise version (1.1) and the common ancestor (1.0). The merged result is version 1.4 in the Enterprise stream.
  10. Propagate the enterprise version 1.4 to the Project 2 stream.
  11. Propagate the project version 1.4 to the Practitioner A stream.
  12. Propagate the project version 1.4 to the Practitioner B stream.
  • Scoping terms
    Scoping is the act of identifying a subset of elements within a larger set. When you scope terms in Information Governance Catalog (IGC), create a category to hold references to all the terms that are part of your scope. The category represents a view of your scope.
  • Scoping data models
    Scoping is the act of identifying a subset of elements within a larger set. When you scope a data model, you create a diagram to hold all the elements that are part of your scope. The diagram represents a view of your scope. Attributes that are part of the scope are highlighted with a color. Attributes that are not part of the scope are set as non-persistent.
  • Propagating data models
    Propagating is the act of disseminating changes to a descendant stream. Propagating is the promotion of changes to one or more of the downstream models such as the enterprise models, the project models or the practitioner models. When propagating a data model, you seed an enterprise data model with the content of an Industry data model, or seed a project data model with the content of an enterprise data model, or seed a practitioner data model with the content of a project data model.
  • Versioning data models
    Versioning is the act of creating a new version of a model with changes. Versioning helps to track and control all changes applied to a model, allowing the recovery of a previous version when needed. When customizing a model, a practitioner creates one or more new versions to capture changes at different stages. A new version of a model can also result from its harvesting from a descendant stream.
  • Harvesting data models
    Harvesting is the act of capturing changes from a descendant stream into the ancestor stream. When harvesting a model, the model owner identifies and reviews model changes. This task consists of comparing and merging the content of a project model into the enterprise model, or the content of a practitioners model into a project model. During harvesting, the model owner accepts or rejects model changes, and resolves potential conflict between the versions. When the latest ancestor version and the ancestor version of the descendant version (common ancestor) are the same, no conflicts exist, and the next ancestor version can be similar to the descendant version, assuming all changes are accepted.
  • Parent topic: Installing the models