Data Warehousing

Data Warehousing

1. Pool of data to support decision making. Application.

   • Subject-oriented, integrated, time-variant, nonvolatille
     • Subject-oriented
       • How and WHY of business performance
     • Integrated: Consistent format
     • Time variant: keeps historical data.
       • Detect trends, deviations, long-term results, etc
     • Nonvolatille: NO CHANGES
   • Web based. Client/server. Relational/multidimensional. Real Time

2. TYPES

   • Data Marts: subset
     • Dependent: directly from DW
     • Independent: small DW, not from an EDW
   • Operational Data Stores (ODS)
     • Constantly updated. Short-term decisions
     • Oper marts: multidimensional analysis. Comes from ODS
   • Enterprise Data Warehouses (EDW)
     • Provide data for Decision Support Systems

3. Process

   • Data Source
   • Extraction and transformation: ETL
   • Data loading
   • Comprehensive database
   • Metadata: data of the data
   • Middleware tools: enable access

4. Architecture

   • Independent Data Marts
   • Bus
   • Hub&Spoke
   • Centralized (No dep. DM)
   • Federated

5. Integration

   • EAI
     • Enterprise Application Integration
   • EII
     • Enterprise Information Integration
   • ETL
     • Extraction, Transformation and Load.