Data Warehouse: architecture and principles
The Data Warehouse it is a data solution to adequately support decision-making processes. After more than twenty years of experience, past choices, both architectural and in terms of data use, need to be reconsidered in light of emerging business models that see 24x7 connection to the information system as a prerequisite. On the Back End, the Cloud relocates the database, while on the Front End, Mobile devices allow a continuously expanding user base to access information. Another important aspect is that of technologies supporting Big Data, whether structured or not. The use of Data Lakes instead of Operational Data Stores or Staging Areas is now common knowledge: when is one better than the other? And what simplification can Data Virtualization bring to the entire architecture? The course, starting from the Data Warehouse lifecycle, explores the new technologies available (DW, ETL, ELT, BI) and new information requests coming from the business (from marketing to management control, to Customer Care, etc.) with the aim of identifying the most appropriate answer in relation to the user's needs. The main architectures are also examined, from the classic 2/3-tier ones, up to the Lambda / Gamma-Delta Architecture, highlighting their characteristics, pros and cons and comparing them in terms of usage needs. The objective of the course is therefore to provide a Comprehensive overview especially from the perspective of data structures, their lifecycles, and Data Governance.
Contents
Data Warehousing Environment Architecture
Comparison of different architectures (Data Warehouse, Data Mart, and ODS), comparison of models (SQL, NoSQL, Star Schema, and derivatives)
What are they, what are the principles, when are they useful, and what are the parameters to keep under control?.
On-Premise and Cloud, with a review of the main market offerings (from Amazon to Snowflake).
ETL, ELT and ESB, through to Data Virtualisation.
Problems and techniques for the construction of software component components.
When and how to integrate them, positioning.
Role within the Data Warehousing, Data Catalog environment.
From Business Glossary to Data Catalog, schema derivation, to summary schemas. Metadata and data quality.
Segmentation and user types.
Comparison between traditional PM approach and Agile approach (requirements, analysis methods, Test strategies).
Prerequisites
Basic knowledge of management and business intelligence systems, data, and the software lifecycle.
Recipients