Establishing 'Making it Happen' as a 'Formal & Predictable' Discipline

DW Dimensional Modeling Process	DW Testing and Implementation

Execution-MiH ENCYCLOPEDIA → Enterprise Intelligence → SECTION - Data-Warehouse/Mart →

CHAPTER -

DW Design & Architecture

This chapter deals with translating 'what' of business model to 'how' of Design & Architecture. Lets look at Extraction, transformation, loading, job control & audit, access services, quality assurance, infrastructure & lot more.

Topics

Data Warehouse Design and Architecture Overview

Data Warehouse Design is the blue print of Extraction, Transformation, loading, source system mapping, Data Warehouse Access & browsing, Data Quality and upstream target systems.

Data Warehouse Source Systems

Identifying the source systems from where the various data elements can be extracted with maximum reliability and minimum effort.

Data Warehouse ETL Extraction

Detailing the extraction of data from Source systems to staging database including the logic, sequence, timings and checks.

Data Warehouse ETL Transformation

Designing the transformation process including standardizing, integrating, cleansing, augmenting, aggregating and creating the data sets for loading into the repository.

Data Warehouse ETL Loading

Loading the data-sets into the data warehouse repository, to ensure that loading happens using minimum system resources and fastest possible time.

Data Warehouse Metadata

All Data Warehouse specific meta data components are listed out and explained.

Back-Room Data Warehouse Metadata

Back-Room Metadata spans across the Data Source and BI Technical Metadata areas and hence occupies a large scope. It encompasses the ETL metadata, data model, security profiles and audit/usage details.

Data Warehouse Data Quality assurance

Data Warehouse operations are mostly through batch processing. Adequate validations are designed to ensure that there is integrity of data through its journey from Source system to DW repository.

Data Warehouse job control and audit

Batch processes are the core to Data Warehouse operations. The design of managing these batch jobs is important.

Data Warehouse sharing and browsing

Cooked data being available in repository now needs to be services to the users through access and browsing services.

Data Warehouse Infrastructure

This page provides the Data Warehouse Infrastructure considerations, which are unique to a Data Warehouse. Otherwise most of the considerations are same as that of any OLTP systems. The unique considerations are mainly linked to the ad-hoc and unpredictable nature of the use a Data Warehouse may be put to.