Table of Contents

• Title and Copyright Information
• Preface
  • Audience
  • Documentation Accessibility
  • Related Documents
  • Conventions
• Changes in This Release for Oracle Database In-Memory Guide
  • Changes in Oracle Database 12c Release 2 (12.2.0.1)
    • New Features
• Part I Oracle Database In-Memory Concepts
  • 1 Introduction to Oracle Database In-Memory
    • 1.1 Challenges for Analytic Applications
    • 1.2 The Single-Format Approach
    • 1.3 The Oracle Database In-Memory Solution
      • 1.3.1 What Is Database In-Memory?
        • 1.3.1.1 IM Column Store
        • 1.3.1.2 Advanced Query Optimizations
        • 1.3.1.3 High Availability Support
      • 1.3.2 Improved Performance for Analytic Queries
        • 1.3.2.1 Improved Performance for Data Scans
        • 1.3.2.2 Improved Performance for Joins
        • 1.3.2.3 Improved Performance for Aggregation
      • 1.3.3 Improved Performance for Mixed Workloads
      • 1.3.4 High Availability Support
      • 1.3.5 Ease of Adoption
    • 1.4 Prerequisites for Database In-Memory
    • 1.5 Principal Tasks for Database In-Memory
    • 1.6 Tools for the IM Column Store
      • 1.6.1 In-Memory Advisor
      • 1.6.2 Cloud Control Pages for the IM Column Store
      • 1.6.3 Oracle Compression Advisor
      • 1.6.4 Oracle Data Pump and the IM Column Store
  • 2 In-Memory Column Store Architecture
    • 2.1 Dual-Format: Column and Row
      • 2.1.1 Columnar Data in the In-Memory Area
        • 2.1.1.1 Size of the In-Memory Area
        • 2.1.1.2 Memory Pools in the In-Memory Area
      • 2.1.2 Row Data in the Database Buffer Cache
    • 2.2 In-Memory Storage Units
      • 2.2.1 In-Memory Compression Units (IMCUs)
        • 2.2.1.1 IMCUs and Schema Objects
          • 2.2.1.1.1 In-Memory Compression
          • 2.2.1.1.2 IMCUs and Rows
        • 2.2.1.2 Column Compression Units (CUs)
          • 2.2.1.2.1 Structure of a CU
          • 2.2.1.2.2 Local Dictionary
        • 2.2.1.3 In-Memory Storage Indexes
      • 2.2.2 Snapshot Metadata Units (SMUs)
        • 2.2.2.1 IMCUs and SMUs
        • 2.2.2.2 Transaction Journal
      • 2.2.3 In-Memory Expression Units (IMEUs)
    • 2.3 Expression Statistics Store (ESS)
    • 2.4 In-Memory Process Architecture
      • 2.4.1 In-Memory Coordinator Process (IMCO)
      • 2.4.2 Space Management Worker Processes (Wnnn)
    • 2.5 CPU Architecture: SIMD Vector Processing
• Part II Configuring the IM Column Store
  • 3 Enabling and Sizing the IM Column Store
    • 3.1 Overview of Enabling the IM Column Store
    • 3.2 Estimating the Required Size of the IM Column Store
    • 3.3 Enabling the IM Column Store for a Database
    • 3.4 Increasing the Size of the IM Column Store Dynamically
    • 3.5 Disabling the IM Column Store
  • 4 Enabling Objects for In-Memory Population
    • 4.1 About In-Memory Population
      • 4.1.1 Purpose of In-Memory Population
      • 4.1.2 How In-Memory Population Works
        • 4.1.2.1 Prioritization of In-Memory Population
        • 4.1.2.2 How Background Processes Populate IMCUs
      • 4.1.3 Controls for In-Memory Population
        • 4.1.3.1 The INMEMORY Subclause
        • 4.1.3.2 Priority Options for In-Memory Population
        • 4.1.3.3 IM Column Store Compression Methods
        • 4.1.3.4 Oracle Compression Advisor
    • 4.2 Enabling and Disabling Tables for the IM Column Store
      • 4.2.1 Enabling New Tables for the In-Memory Column Store
      • 4.2.2 Enabling and Disabling Existing Tables for the IM Column Store
      • 4.2.3 Enabling and Disabling Tables for the IM Column Store: Examples
    • 4.3 Enabling and Disabling Columns for In-Memory Tables
      • 4.3.1 About IM Virtual Columns
      • 4.3.2 Enabling IM Virtual Columns
      • 4.3.3 Enabling a Subset of Columns for the IM Column Store: Example
      • 4.3.4 Specifying INMEMORY Column Attributes on a NO INMEMORY Table: Example
    • 4.4 Enabling and Disabling Tablespaces for the IM Column Store
    • 4.5 Enabling and Disabling Materialized Views for the IM Column Store
    • 4.6 Forcing Initial Population of an In-Memory Object: Tutorial
    • 4.7 Enabling ADO for the IM Column Store
      • 4.7.1 About ADO Policies and the IM Column Store
      • 4.7.2 Purpose of ADO and the IM Column Store
      • 4.7.3 How ADO Works with Columnar Data
        • 4.7.3.1 How Heat Map Works
        • 4.7.3.2 How Policy Evaluation Works
      • 4.7.4 Controls for ADO and the IM Column Store
      • 4.7.5 Creating an ADO Policy for the IM Column Store
• Part III Optimizing In-Memory Queries
  • 5 Optimizing Queries with In-Memory Expressions
    • 5.1 About IM Expressions
      • 5.1.1 Purpose of IM Expressions
      • 5.1.2 How IM Expressions Work
        • 5.1.2.1 IM Expressions Infrastructure
        • 5.1.2.2 Capture of IM Expressions
        • 5.1.2.3 How the ESS Works
        • 5.1.2.4 How the Database Populates IM Expressions
        • 5.1.2.5 How IMEUs Relate to IMCUs
      • 5.1.3 User Interfaces for IM Expressions
        • 5.1.3.1 INMEMORY_EXPRESSIONS_USAGE
        • 5.1.3.2 DBMS_INMEMORY_ADMIN and DBMS_INMEMORY
      • 5.1.4 Basic Tasks for IM Expressions
    • 5.2 Configuring IM Expression Usage
    • 5.3 Capturing and Populating IM Expressions
    • 5.4 Dropping IM Expressions
  • 6 Optimizing Joins with Join Groups
    • 6.1 About In-Memory Joins
    • 6.2 About Join Groups
    • 6.3 Purpose of Join Groups
    • 6.4 How Join Groups Work
      • 6.4.1 How a Join Group Uses a Common Dictionary
      • 6.4.2 How a Join Group Optimizes Scans
      • 6.4.3 When a Hash Join Uses Common Dictionary Encodings
    • 6.5 Creating Join Groups
    • 6.6 Monitoring Join Group Usage
      • 6.6.1 Monitoring Join Groups Using a SQL Monitor Report: Example
      • 6.6.2 Monitoring Join Groups from the Command Line: Example
  • 7 Optimizing Joins with In-Memory Aggregation
    • 7.1 About IM Aggregation
    • 7.2 Purpose of IM Aggregation
      • 7.2.1 When IM Aggregation Is Useful
      • 7.2.2 When IM Aggregation Is Not Beneficial
    • 7.3 How In-Memory Aggregation Works
      • 7.3.1 When the Optimizer Chooses IM Aggregation
      • 7.3.2 Key Vector
      • 7.3.3 Two Phases of IM Aggregation
      • 7.3.4 IM Aggregation: Scenario
        • 7.3.4.1 Sample Analytic Query of a Star Schema
        • 7.3.4.2 Step 1: Key Vector and Temporary Table Creation for geography Dimension
        • 7.3.4.3 Step 2: Key Vector and Temporary Table Creation for products Dimension
        • 7.3.4.4 Step 3: Key Vector Query Transformation
        • 7.3.4.5 Step 4: Row Filtering from Fact Table
        • 7.3.4.6 Step 5: Aggregation Using an Array
        • 7.3.4.7 Step 6: Join Back to Temporary Tables
    • 7.4 Controls for IM Aggregation
    • 7.5 In-Memory Aggregation: Example
  • 8 Optimizing Repopulation of the IM Column Store
    • 8.1 About Repopulation of the IM Column Store
    • 8.2 How Data Loading Works with the IM Column Store
      • 8.2.1 How Conventional DML Works with the IM Column Store
        • 8.2.1.1 Staleness Threshold
        • 8.2.1.2 Double Buffering
      • 8.2.2 How Direct Path Loads Work with the IM Column Store
      • 8.2.3 How a Partition Exchange Load Works with the IM Column Store
    • 8.3 When the Database Repopulates the IM Column Store
      • 8.3.1 Threshold-Based and Trickle Repopulation
      • 8.3.2 Factors Affecting Repopulation
    • 8.4 Controls for Repopulation of the IM Column Store
    • 8.5 Optimizing Trickle Repopulation: Tutorial
• Part IV High Availability and the IM Column Store
  • 9 Managing IM FastStart for the IM Column Store
    • 9.1 About IM FastStart
      • 9.1.1 Purpose of IM FastStart
      • 9.1.2 How IM FastStart Works
        • 9.1.2.1 How the Database Manages the FastStart Area
        • 9.1.2.2 How the Database Reads from the FastStart Area
    • 9.2 Enabling IM FastStart for the IM Column Store
    • 9.3 Retrieving the Name of the Current IM FastStart Tablespace
    • 9.4 Migrating the FastStart Area to a Different Tablespace
    • 9.5 Disabling IM FastStart for the IM Column Store
  • 10 Deploying IM Column Stores in Oracle RAC
    • 10.1 Overview of Database In-Memory and Oracle RAC
      • 10.1.1 Multiple IM Column Stores
      • 10.1.2 Distribution and Duplication of Columnar Data in Oracle RAC
        • 10.1.2.1 Distribution of Columnar Data in Oracle RAC
          • 10.1.2.1.1 Distribution by Partition
          • 10.1.2.1.2 Distribution by Subpartition
          • 10.1.2.1.3 Distribution by Rowid Range
        • 10.1.2.2 Duplication of Columnar Data in Oracle RAC
          • 10.1.2.2.1 DUPLICATE Clause in Oracle RAC
          • 10.1.2.2.2 DUPLICATE ALL Clause in Oracle RAC
          • 10.1.2.2.3 NO DUPLICATE Clause in Oracle RAC
      • 10.1.3 Parallelism in Oracle RAC
        • 10.1.3.1 Serial and Parallel Queries in Oracle RAC
        • 10.1.3.2 Auto DOP in Oracle RAC
      • 10.1.4 FastStart Area in Oracle RAC
    • 10.2 Configuring In-Memory Services in Oracle RAC
      • 10.2.1 Instance-Level Service Controls
      • 10.2.2 Object-Level Service Controls
      • 10.2.3 Benefits of Services for Database In-Memory in Oracle RAC
      • 10.2.4 Configuring an In-Memory Service for a Subset of Nodes: Example
  • 11 Deploying an IM Column Store with Oracle Active Data Guard
    • 11.1 About Database In-Memory and Active Data Guard
      • 11.1.1 Purpose of IM Column Stores in Oracle Active Data Guard
        • 11.1.1.1 Identical IM Column Stores in Primary and Standby Databases
        • 11.1.1.2 IM Column Store in Standby Database Only
        • 11.1.1.3 Different Objects in the Primary and Standby IM Column Stores
      • 11.1.2 How IM Column Stores Work in Oracle Active Data Guard
    • 11.2 Configuring IM Column Stores in an Oracle Active Data Guard Environment
• Part V Database In-Memory Reference
  • 12 In-Memory Initialization Parameters
  • 13 In-Memory Views
• A Using IM Column Store in Cloud Control
  • A.1 Meeting Prerequisites for Using IM Column Store in Cloud Control
  • A.2 Using the In-Memory Column Store Central Home Page to Monitor In-Memory Support for Database Objects
  • A.3 Specifying In-Memory Details When Creating a Table or Partition
  • A.4 Viewing or Editing IM Column Store Details of a Table
  • A.5 Viewing or Editing IM Column Store Details of a Partition
  • A.6 Specifying IM Column Store Details During Tablespace Creation
  • A.7 Viewing and Editing IM Column Store Details of a Tablespace
  • A.8 Specifying IM Column Store Details During Materialized View Creation
  • A.9 Viewing or Editing IM Column Store Details of a Materialized View
• Glossary
• Index