Participants will gain an understanding of Oracle’s database partitioning architecture and learn how partitioning can benefit data management. They will also explore how parallel operations can reduce response time for data-intensive tasks. The course covers extracting, transforming, and loading data into an Oracle data warehouse, as well as using materialized views to enhance performance. Additionally, participants will learn how query rewrites can further improve performance in a data warehouse environment.

Introduction

• Development tools
• Oracle SQL Developer
• Enterprise Manager
• Sample schemas used

Data Warehouse Design: Overview

• Characteristics of a Data Warehouse
• Comparing OLTP and Data Warehouses
• Data Warehouse Architectures
• Data Warehouse Design and Objects
• Data Warehouse Schemas

Data Warehouse Tuning Considerations

• Optimizing Star Queries
• Bitmap Join Indexes
• Star Query Optimization and Bitmap Join Index Optimization

Partitioning Basics

• Partitioned Tables and Indexes
• Partitioning Methods and Types
• Partition Pruning and Star queries

Parallelism Concepts

• Operations that can be parallelized
• Parallel execution mechanics
• Degree of parallelism
• Parallel execution plan
• Automatic parallelism

Parallel Operations in Data Warehouses

• Parallel Query, DDL, and DML
• Tuning parallel execution parameters
• Balancing the workload

ETL: Extraction and Transportation

• Extraction methods
• Capturing data with Change Data Capture (CDC)
• Sources and modes of Change Data Capture  (CDC)
• Publish and Subscribe models: The Publisher and the Subscriber
• Synchronous and Asynchronous CDC
• Asynchronous AutoLog mode and Asynchronous HotLog mode
• Transportation in a data warehouse
• Transportable tablespaces

ETL: Loading

• Loading mechanisms
• Applications of external tables
• Defining external tables with SQL*Loader
• Populating external tables with Data Pump
• Other loading methods

ETL: Transformation

• Data transformation
• Transformation mechanisms
• Transformation using SQL
• Table functions
• DML error logging

Materialized Views

• The need for summary management
• Types of materialized views
• Using materialized views for summary management
• Materialized view dictionary views

Refreshing Materialized Views

• Refresh options and modes
• Conditions affecting fast refresh
• Materialized view logs and PCT refresh
• Refresh performance improvements

Working With Dimensions

• What are dimensions
• Creating dimensions and hierarchies
• Dimensions and privileges
• Dimension restrictions
• Verifying relationships in a dimension
• Dimension invalidation

Query Rewrite

• Query Rewrite: An overview
• What can be rewritten
• Conditions required for Oracle to rewrite a query
• Query Rewrite guidelines
• Setting initialization parameters for Query Rewrite
• Query Rewrite methods
• Partition Change Tracking (PCT) and Query Rewrite
• Query Rewrite enhancement to support queries containing inline views

Using the SQL Access Advisor, Compression, and Resumable Sessions

• SQL Access Advisor: Usage Model
• Setting initial options
• Specifying the Workload Source
• Recommendation options
• Schedule and Review
• PL/SQL procedure flow
• Tuning materialized views for Fast Refresh and Query Rewrite
• Table compression and resumable sessions

You will learn how to implement partitioning, use parallel operations to reduce response time, perform ETL (Extract, Transform, Load) processes, create and manage materialized views to enhance data warehouse performance, leverage query rewriting with materialized views for faster business queries, and use SQL Access Advisor and PL/SQL procedures to optimize materialized views for fast refresh and query rewriting.