Recognises an entity split into clone tables by value
One entity stored as several tables with the same columns, split by a value: a year, a region, a status. Uniqueness and foreign keys cannot span them, a change to the split value means moving the row, and every question across values needs a union. The physical form of logical row 28.
Can AI do this?
Not tested yet
No AI benchmark covers this knobit yet. New results are announced in the Can AI Do the Job? newsletter.
What it looks like in practice
Expected result: The tables and the value to make a column
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: payment_2025, payment_2026; payment_identifier unique in each but not across both; payment_hold cannot reference "a payment"; a corrected value date moves the payment between tables Acceptable: One payment table partitioned by payment_value_date in physical_options, keys and foreign keys unchanged. Where separate tables are unavoidable (an archive), their predicates are disjoint and written down, and uniqueness across them is a DQ rule
Also known as: Horizontal split into clone tables (physical model)
Related knobits
Builds on
- K-528 Recognises one entity split by value into several
- K-660 Explains vertical and horizontal partitioning of tables
All 40 knobits of Design the physical data model.
Source: Review checklist 06-physical, row 5; Karwin, Metadata Tribbles; Date & McGoveran 1994; S&W 12.6