Skip to main content

Semantic Model

Deep dive into Strata's semantic modeling capabilities.

Overview​

The semantic model is the core of Strata. It transforms raw database tables into business-friendly analytics models that enable self-service analytics.

Critical Rules​

Before diving into components, understand these non-negotiable design principles:

Unique Naming Principle — Strata's Approach

Every field name must be unique across your entire semantic layer.

This is a fundamental design principle (unlike tools that use namespaced fields like table.field_name).

Benefits:

  • Unambiguous references: "Total Revenue" means exactly one thing
  • Compound measures: [Total Revenue] - [Total Cost] works without table prefixes
  • AI integration: LLMs can reference fields without disambiguation
  • Consistent metrics: Organization-wide definitions

Considerations:

  • Requires thoughtful naming conventions from the start
  • Use prefixes or suffixes to avoid conflicts (e.g., order_revenue, subscription_revenue)
# These cannot coexist in the same project:
# In tbl.store_sales.yml
- name: Total Revenue  # ✓ First definition

# In tbl.catalog_sales.yml  
- name: Total Revenue  # ✗ ERROR: Name already used
- name: Catalog Revenue  # ✓ Use a unique name instead
No Many-to-Many Relationships

many_to_many cardinality is not supported. This design constraint prevents double-counting bugs and ensures predictable aggregations.

Workaround: Use junction tables with two separate relationships.

# ✗ NOT SUPPORTED
users_roles:
  cardinality: many_to_many

# ✓ USE JUNCTION TABLE
users_user_roles:
  left: Users
  right: User Roles
  cardinality: one_to_many

user_roles_roles:
  left: User Roles
  right: Roles
  cardinality: many_to_one

Why: This approach makes relationships explicit and prevents common aggregation errors where measures get counted multiple times.

Key Components​

Tables​

Tables define semantic metadata for physical database tables:

  • Dimensions: Categorical fields for grouping
  • Measures: Aggregatable metrics
  • Cost: Query optimization hints
  • Partitions: Data availability constraints

Learn about tables →

Fields​

Fields are the building blocks of your semantic model:

  • Dimensions: Attributes for grouping and filtering
  • Measures: Metrics for aggregation
  • Data Types: string, integer, decimal, date, etc.
  • Field format: Currency, percent, dates, and display formatting

Learn about fields →

Expressions​

Expressions define how fields query the database:

  • SQL: Direct SQL expressions
  • Lookups: Dimension field indicators
  • Arrays: Multi-value fields
  • Primary Keys: Unique identifiers

Learn about expressions →

Relationships​

Relationships define how tables join:

  • Cardinality: one-to-one, one-to-many, many-to-one
  • Join Types: inner (default), left, right
  • Join Conditions: SQL expressions

Learn about relationships →

How It Works​

  1. Model Definition: Data engineers define tables and relationships in YAML
  2. Deployment: Models are deployed to Strata server
  3. Universe Formation: Planner generates all possible query paths
  4. Query Planning: User queries are converted to optimized SQL
  5. Execution: SQL runs against datasources
  6. Results: Formatted results returned to users

Advanced (in-depth and extra)​

The Advanced subsection below covers deeper and optional semantic-model features: exclusions, inclusions, partitions, decorators, extended blending, cost optimization, and multi-datasource. The core basics (tables, fields, expressions, relationships, imports) are above; Advanced is for more in-depth and extra scenarios.

  • Imports: Reuse field definitions across tables
  • Partitions: Define data availability constraints
  • Snapshot Measures: Point-in-time analysis
  • Exclusions: Filter-based measure exclusions
  • Inclusions: Multi-level calculations
  • Decorators: Temporal, window, contribution functions

Advanced features →

Next Steps​