Skip to content
Software Architecture Guide

Data Entity/Data Component Catalog

Explanation

Section titled “Explanation”

A data entity is a kind of business data, such as a customer or an order. The catalog is the master inventory of all such entities in the enterprise. For each one it records a definition, an owner and a classification level.

System of record. Each entity should have exactly one designated system of record. This is the authoritative source that creates and updates the entity. Other systems may use the data, but they do not own it.

Owner and steward. The owner holds strategic accountability for the entity. A steward is a different role. Stewards handle day-to-day data quality.

Classification. The catalog states how sensitive each entity is. This level drives encryption, access control, retention and compliance requirements. A sensitive entity therefore gets stricter handling than an ordinary one.

Links to other artifacts.

  • A Data Entity/Data Component Matrix maps logical entities to the physical storage systems that hold them.
  • A CRUD matrix maps application components to data entities. It declares which application holds authority over each entity. (CRUD stands for create, read, update, delete.)
  • TOGAF-aligned data governance work can include entity and component catalogs, with entities mapped to concrete tables and views. See the article on TOGAF.

Neighboring tools. A business glossary defines business terms and is separate from a catalog of data assets. A data dictionary gives technical information about data assets and elements. A data catalog helps users judge which data assets exist and how they are classified. The entity catalog works alongside these.

Keeping it current. DAMA-DMBOK practices help keep entities, definitions, metadata, quality rules and stewardship up to date.

Example

Section titled “Example”

A retail company lists “Customer” in its catalog. The entry gives a short definition and names a business owner and a steward. It marks the entity as sensitive. It names the customer management application as the single system of record. When a new web shop needs customer data, the team reads the entry. They learn who to ask, how strictly to protect the data, and which system is authoritative.

Trade-offs

Section titled “Trade-offs”

A catalog only helps while it is accurate. Definitions, owners and stewardship change, so someone must keep them current. Insisting on a single system of record also means every other system must treat that source as authoritative for the entity.

Checks

Section titled “Checks”
  • Every data entity in the catalog has a definition, an owner and a classification level.
  • Each entity names exactly one system of record.
  • Owner and steward roles are distinguished for each entity.
  • Classification levels are linked to handling requirements such as encryption and access control.
  • Entities are mapped to the physical storage systems or components that hold them.
Section titled “Related”

Sources

Section titled “Sources”