🗓️ 27052025 1356
📎

data_warehouse_naming_conventions

Common Layer Prefixes

Prefix	Meaning	Usage
`ods_`	Operational Data Store	Raw, unprocessed data ingested from sources (e.g., logs, Kafka, RDS dumps). Often partitioned by day/hour.
`dwd_`	Data Warehouse Detail	Cleansed and detailed data, usually 1:1 with business events/entities. Normalized and enriched.
`dws_`	Data Warehouse Summary	Aggregated data for specific business domains. Used for reporting, metrics, dashboards.
`ads_`	Application Data Store	Final layer used directly by applications or APIs (reports, dashboards, downstream services).
`dim_`	Dimension Table	Lookup tables with descriptive attributes (e.g., `dim_user`, `dim_region`).
`tmp_`	Temporary Table	Intermediate tables used in ETL processing. Not meant for long-term use.
`rpt_`	Report Table	Tables generated specifically for business reports, often updated on schedule.

Suffix / Tag	Meaning	Notes
`_di`	Daily Incremental	Updated once per day, may include only changes (upserts/inserts).
`_fi`	Full Incremental	Full daily snapshot, usually for slowly changing dimensions or syncs.
`_ri`	Real-Time Incremental	Updated in near-real-time, often with Flink/Kafka pipelines.
`_hi`	Historical Incremental	Stores historical data with full SCD (slowly changing dimension) support. Usually partitions over long time spans.
`_full`	Full Snapshot	Typically refers to a full data snapshot without incremental logic.
`_bak` / `_backup`	Backup Table	Used for temporary backups, often before destructive operations.