Differences from Traditional Warehouse Architecture

3.3 Differences from Traditional Warehouse Architecture

This table summarizes the differences of the OFSDF from a traditional Data Warehouse architecture.

Table 3-1 Differences between the traditional Data Warehouse architecture and OFSDF

Component	FSDF	Traditional BI Architecture
Staging Layer	Common staging Area (CSA) where data from source systems are staged and is implemented as database schema.	Usually a file system-based area where file-based extracts (operational images) from source systems are staged prior to loading into a target schema.
3rd Normal Form Operational Data Store/Enterprise Data Warehouse	Does not provide a physicalized model for a 3NF store. Operational/fine-grained reporting will be fulfilled from the reporting area.	Notification sent to all users captured as owners and user who submitted the definition.
Data Marts/Reporting Model	Set of star schemas with conformed dimensions (Ralph Kimball approach)	Set of star schemas

Component

FSDF

Traditional BI Architecture

Staging Layer

Common staging Area (CSA) where data from source systems are staged and is implemented as database schema.

Usually a file system-based area where file-based extracts (operational images) from source systems are staged prior to loading into a target schema.

3rd Normal Form Operational Data Store/Enterprise Data Warehouse

Does not provide a physicalized model for a 3NF store.

Operational/fine-grained reporting will be fulfilled from the reporting area.

Notification sent to all users captured as owners and user who submitted the definition.

Data Marts/Reporting Model

Set of star schemas with conformed dimensions (Ralph Kimball approach)

Set of star schemas

With the preceding understanding in mind, the following sections describe the data models in the OFSDF in greater detail.

Readily Deployable: The Physical Data Model is a readily deployable physical schema. It is provided as an erwin Data Model file (for details on erwin, seehttps://erwin.com/products/erwin-data-modeler/) and consists of tables grouped into distinct subject areas depending on function. The tables are either used to gather source data (Staging Area), or as containers of outputs/results from analytical processing and engines for reporting purposes (Reporting Area).
Use-case Driven: The OFSDF Physical Data model is driven by a set of clearly identified analytical use cases spanning Risk, Performance, Customer Insight, and Compliance.
Extensible: While the OFSDF Physical Data Model satisfies a very large number of analytical use cases across Risk, Finance, Marketing, and Compliance subject areas, customers may find the need to customize the model for a specific installation.

These customizations may be done in accordance with guidelines published in Using OFSDF section of this manual.

The OFSDF Physical Data Model is divided into two primary areas: