TwoTier DW Architecture ThreeTier DW Architecture DW Components

DW Components • Data migration tools – Tools that help extract, transform and load

DW Components • Warehouse data stores – Structures used to actually store the data

Steps in Data Warehousing 1. 2. 3. 4. Identify all the sources of data

DW Design: Star Schema Lots of records, but each record is “thin” “Fact” Want

E/T/L Process • Extract – Data must be extracted from source systems • Transform

Slides: 8

Download presentation

Two-Tier DW Architecture

Three-Tier DW Architecture

DW Components • Data migration tools – Tools that help extract, transform and load data into the data warehouse – Three main categories • Data copying and replication • Data transformation • Data cleansing • Metadata usage – Administrative and end-user

DW Components • Warehouse data stores – Structures used to actually store the data – Typically relational DB (but not always) – Multi-dimensional DB becoming more popular • Data retrieval, formatting and analysis – Query tools – Analysis tools – Data mining • Management tools – Access control, performance monitoring, usage monitoring

Steps in Data Warehousing 1. 2. 3. 4. Identify all the sources of data Design the data warehouse Extract/Transform/Load process Decision making from DW

DW Design: Star Schema Lots of records, but each record is “thin” “Fact” Want to be able to “see” each Sale by Product, Time, Store “Dimensions” Fewer records, but each record is “fat” (lots of big columns)

E/T/L Process • Extract – Data must be extracted from source systems • Transform – Cleansing • Identify and eliminate data inconsistencies • Can be very complex, expensive, time consuming – Aggregation • Load – Must be repeated periodically – How often? – Identifying changed data

DW Process Overview