Arivo Data Vault Entity Models

Design Objectives

Entity model considerations will prioritize agility over efficiency optimizations. Efficiency will play a role in the design, but some sacrifices to join optimizations will be made in order to ensure that model evolution can take place easily without large re-engineering costs to adapt to shifting business needs. Dimensional modeling leads to rigid table schemas fit to purpose that often lead to massive re-engineering requirements should the business data needs shift. These designs will attempt to emulate the Data Vault 2.0 as laid out in Building Scalable Data Warehouse with Data Vault 2.0 (Linstet and Olschmike, 2016). This design approach is less rigid and provides a more flexible schema layout that can be appended to easily to expand the data warehouse model for the business. Generally, this model requires more joins and thus slightly more complex SQL statements.

Application

LoadDate - Timestamp indicating when the data was loaded into the warehouse
ExtractDate - Timestamp indicating when the record was loaded into the staging area RecordSource - Indicates from which source the record was loaded from

Satellites

Satellites in this model capture change data. Eventually this model should be split into multiple tables, where faster changing data, i.e. status, may be captured and reduce replication of the data that rarely or never changes from initial creation. The composite will then be the ApplicationHashKey and the UpdatedAt keys.

Borrower Hub

Under Development, expected Feb 19th.

Satellites

Borrower Incomes
Primary borrower income table as determined by application logic from various employment verification sources.

Borrower Employments
Borrower supplied income information.

Borrower Employments Verification
Income verification as reported by various services for a given borrower.

Dealer Hub

Currently Under Development, expected Feb 26th.

Account Hub

This will wait until account servicing components will begin development. Expected in Q2 2021.