What's the Buzz About Composite Entities?
In many scenarios, a linear dimension does the trick for scoping controls. Take, for instance, a control affecting a business application:
- Admin access from outside the corporate network must be disabled – SAP
However, for larger organizations, scoping controls by multiple dimensions becomes essential. Enter ‘Composite Entities’. These allow you to take several tables (dimensions) of master data and 'multiply' them out for each control—a sophisticated yet elegant solution.
High-Level Steps to Use Composite Entities
Consider this example combining business processes and locations.
Load Master Data
- Enter your list of locations and business processes in to the appropriate foundation or CMDB tables.
Create an Entity Type for Each Dimension
- Establish Entity Types for 'Location' and 'Business Process'.
- Pro Tip: Don’t forget to click ‘Update Entities from Filters‘ to generate the entities straight away.
Create a New Composite Entity Class
- Select the two dimensions previously created to create a new Class.
- Select the two dimensions previously created to create a new Class.
- Create New Entities Within This Composite Class:
- For each entity, select one from each dimension.
- Heads-Up: You'll need to repeat this step for every valid combination of entities in your environment.
Closing Thoughts
One significant advantage of the Composite Entities feature is that it automatically builds the upstream dependency between the composite entity and the underlying entity.
One limitation I’ve encountered is the lack of an out-of-the-box mechanism to build upstream entities based on the ‘parent’ field of the underlying master data. It would be fab if the location and business process) hierarchies were automatically mapped out as upstream entities. There is a similar need when using CMDB CIs with relationships.
Have you had the chance to use Composite Entities yet? I’d love to hear your thoughts and experiences. Sound off in the comments!
