What are data models?

Data models provide a framework to describe how data is structured, organized, and related within a system. It acts as a blueprint for organizations to design their business applications and processes. Data models can be of different types: relational, hierarchical, entity relationship, and network.

Atlan enables you to ingest your entity–relationship (ER) models and associate them with existing data assets in Atlan. Cataloging your ER model metadata in Atlan can help you:

• Foster collaboration — business and technical users work best when they share a common understanding of the data landscape without tool boundaries.
• Handle change management through impact analysis — data models enable visualization of an asset's lifecycle within an organization, helping users assess business impact due to technical changes with accuracy and vice versa.
• Implement data governance — define access control mechanisms, data retention policies, and data governance rules spanning different systems by understanding relationships between data assets. When business-approved data models are coupled with technical objects, trust and accountability are established between key stakeholders.

Ingest ER models

You can ingest your ER models in Atlan using the following methods:

• Data model ingestion — Atlan recommends using this custom package to ingest your ER models via an Excel template.
• Atlan SDK
• Atlan REST API

Entity–relationship models

Entity–relationship (ER) models focus on entities (objects/concepts) and the attributes (characteristics) and relationships (associations) between those entities.

In the context of entity–relationship modeling, a model encompasses the entities, attributes, and relationships that define how data is organized and interactions between different elements within a specific domain.

Data models can be used to represent information at different levels of abstraction:

• Conceptual — overall structure of content without specific details. This acts as a starting point for new data initiatives and is the most abstract form of the model.
• Logical — implementation-agnostic breakdown of data into specific objects and interactions between these objects.
• Physical — a refined adaptation of data concepts conforming to a particular software application or data storage system. This level takes into account finer nuances like naming conventions, optimizations, partitioning, and more.

Entity-relationship diagrams

An entity-relationship diagram (ERD) is a visual representation of data that illustrates the entities (objects or concepts) within a system, relationships between those entities, and their attributes.

• Entity — in an ERD, an entity is a fundamental component that represents a real-world object or concept within a database. For example, entities are typically nouns, such as Customer, Order, or Product and data can be stored about them.
• Attribute — an entity has attributes, which are the properties or characteristics of the entity. For example, a Customer entity may have attributes like CustomerID, Name, Email, and Phone Number.
• Relationship — a relationship determines how two entities interact with each other. For example, a Customer places an Order. A relationship encompasses several elements, like:
  • Cardinality — defines the quantitative aspect of a relationship. For example, a Quote provides pricing for many related Orders (one-to-many).
  • Optionality — defines whether a relationship is mandatory in an entity. For example, an Order must have an associated Customer.
  • Cardinality and optionality can be combined to define business rules. For example, in a Library system, a Member can borrow 0-n book(s).
  • Types of relationships:
    • Association — refers to a peer-to-peer relationship between two entities.
    • Generalization — refers to a parent-child relationship between two entities. For example, a Loan entity can be of type Home Loan, Auto Loan, Business Loan, and so on.
• Model — in the context of ER modeling, a model encompasses the entities, attributes, and relationships that define how data is organized and how different elements interact within a specific domain.

  • Models can be of different types — conceptual, logical, and physical.

  • Mapping — entities within a model can be mapped to entities within another model of a different type. For example, a logical entity Order can be mapped to your assets in Atlan, such as an Order table in Snowflake.