K201 Sprint 2 Exam Terms

Enforced Foreign Key

constraint where the database actively rejects any record whose foreign key does not exist in the primary key table

Logical Foreign Key

A field documenting a relationship in the schema without active enforcement by the database engine

Referential Integrity

A state where every foreign key reference successfully matches a valid primary key record in the related tableA state where every foreign key reference successfully matches a valid primary key record in the related table

Verification Gap

The difference between what a database appears to be doing (documented rules) and what it actually enforces

Business Rule Violation

Values that are structurally valid for their data type but operationally incorrect (e.g., a review rating of 10 on a 1–5 scale)

Data Quality Failure

Data records that are incomplete, inconsistent, or impossible due to missing automated or human checks

Structured Data

Information organized according to consistent fields, defined data types, and established relational rules

Semi-structured Data

Data organized through labels or keys where fields may repeat, nest, or vary among individual records (e.g., JSON)

Unstructured Data

Content like free text, audio, images, or video that does not arrive in a consistent set of fields

JSON (JavaScript Object Notation)

A text format for app and web data using labeled fields, nested structures, and variable records

Document-Style

A storage home designed specifically for semi-structured records containing nested, repeating, or variable fields

Restricted Repository

A controlled storage environment for current text or staff notes that require restricted access or limited approved uses

Data Lake

A shared repository that stores varied raw data sources together close to their original form before reshaping

Data Warehouse

A platform holding prepared, structured data organized specifically for recurring reports and standardized analysis

Data Architecture

The overarching plan detailing how a business stores its data, connects its components, and makes information available for use

Derived Table

A new reporting structure created from existing source data to summarize, calculate, or combine fields for analysis

Grain

The exact level of detail representing what one single row or record represents in a table or dataset

Provenance

Information tracking where data originated or which system/role produced it

Traceability

The ability to trace a derived summary, field, or score back to its original source records and calculation logic

Reporting Layer

A dedicated architectural layer sitting on top of source systems that organizes validated data for recurring business reporting

CORE Framework

A structured method for writing generative AI architecture briefs covering Context, Objective, Role, and Expectations