Range Partitioning
- Partitions data based on ranges of a key column (e.g., dates, IDs)
- Excellent for time-series data where newer data is accessed more frequently
- Example: Monthly partitions for transaction data (Jan 2025, Feb 2025, etc.)
List Partitioning
- Assigns rows to partitions based on discrete values in a column
- Good for categorical data with a known set of values
- Example: Partitioning sales data by country or region code
Hash Partitioning
- Uses a hash function on the partition key to determine partition placement
- Distributes data evenly across partitions
- Ideal when there’s no natural range or list partitioning strategy
Composite Partitioning
- Combines multiple partitioning methods (e.g., range-hash, list-hash)
- Provides more granular control over data distribution
- Example: First partition by date range, then hash by customer ID