What characterises a star schema?

One fact table with one denormalised table per dimension

Multiple fact tables sharing dimensions

Compared with star, a snowflake schema has:

Less redundancy and storage but more joins

More redundancy and fewer joins

Multiple fact tables sharing dimension tables

Item split into Item and Supplier tables linked by supplier_key

Item alone, with supplier_type inside

Compare Star, Snowflake and Fact Constellation schemas with examples.

Point	Star schema	Snowflake schema	Fact constellation
Structure	One fact table, one table per dimension	Star with normalised (split) dimensions	Multiple fact tables sharing dimensions
Dimension tables	Denormalised	Normalised into sub-tables	Shared between facts
Redundancy	High	Low	Depends on dimensions
Joins	Few	More	Many, across facts
Query speed	Fastest	Slower	Slowest, most complex
Storage	More	Less	More
Complexity	Simple	Moderate	Complex
Use	Data marts, simple models	Large dimension hierarchies	Enterprise warehouses

Star example

 Time    Item
   \    /
  SALES FACT -- Branch

Item(item_key, name, brand, type, supplier_type) is a single table.

Snowflake example

 SALES FACT -- Item -> Supplier
          \
           Location -> City

Item is split into Item(item_key, name, brand, type, supplier_key) and Supplier(supplier_key, supplier_type).

Fact constellation example

 Sales Fact      Shipping Fact
     \            /
   Time  Item  Location  (shared)

Sales and Shipping facts both use the Time, Item and Location dimensions.

✓ Answer checked by Claude Opus 5.5 (high effort)