Information systems and data management
The Data Backbone
Structure, store, and check the data: primary and foreign keys relate the tables, and the quality dimensions decide whether it can be trusted.
How the exam words it
- -The stem defines a key and asks whether it is a primary key or a foreign key, or how referential integrity is enforced.
- -It contrasts storage repositories and asks which is a data warehouse, a data lake, a data mart, or an operational database.
- -It describes duplicated or conflicting records and asks which process addresses it, pointing to master data management.
- -It asks what normalization accomplishes or which quality dimension (completeness, consistency, accuracy, timeliness) is at issue.
The playbook
- 1Define keys by role: a primary key uniquely identifies each row, and a foreign key references a primary key in another table to enforce referential integrity.
- 2Match the repository: a data warehouse holds structured, cleansed data for analysis, a data lake holds raw data of any format, and a data mart is a subject-specific slice.
- 3Use master data management to maintain one authoritative version of core entities (customers, vendors) across systems, reducing duplication and conflict.
- 4Read the quality dimension precisely: completeness means nothing is missing, consistency means no conflicts across sources, accuracy means correctness, and normalization removes redundancy.
The trap
Confusing consistency with completeness, or a data lake with a warehouse. Consistency means no conflicting values across sources; completeness means no missing values.
How the exam varies it
The same pattern, re-skinned along these axes:
Primary versus foreign key and referential integrityWarehouse versus lake versus mart repositoryWhich data-quality dimension, and the goal of normalization
Drill this pattern
8 questions of The Data Backbone from across the AUD topics. Clear it by getting 5 right with a streak of 3.