Preface
Data Replication Overview
- Product Overview
- Data Replication Usage Scenarios
- Data Replication Sources and Targets
- Data Replication Components
- Data Replication Deployment Topologies
- Data Replication Apply Modes
Installing Data Replication
- Installation Overview
- System Requirements
- Installing Data Replication
- Installing the Java Runtime Environment
- JDBC Drivers
- Defining the DBSYNC_HOME Environment Variable
- Adding DBSYNC_HOME Libraries to the Library Path Environment Variable on Linux and UNIX
- Verifying the Environment Variables on Linux and UNIX
- Setting Limits for System Resources on Linux and UNIX
- Entering the License Key
Preparing Source and Target Systems
- Source and Target Preparation and Replication Considerations
- ODBC Drivers
  - Supported ODBC Drivers
  - Configuring the DataDirect ODBC Drivers and Driver Manager
- Preparing Source Systems
- Preparing Target Systems
- Replication Configuration Considerations for Sources
- Replication Configuration Considerations for Targets
Replication Basic Concepts
- Understanding Replication Basics
- Strategies for Running Replications
- Replication Configurations
- Database Character Set Conversion
- Replication of Database-Generated Values
- Database Logging Requirements for Source Tables
- Replicating Tables without a Primary Key Definition
- Datatype Conversion Rules
- Tcl and SQL Scripting for Advanced InitialSync and Applier Processing of Data
- DBSYNC_SYNC_LSN Table
InitialSync Processing
- Target Materialization with InitialSync
- Source Connectivity for Data Unload Operations
- Target Connectivity for Data Load Operations
- Sync SCN Value
- InitialSync Handling of Target Table Constraints
- Considerations for Running InitialSync
Extractor and Applier Processing
- Start SCN Value for the Extractor Task
- Extractor Processing of Redo Logs from Oracle RAC Sources
  - The Order of Writing Extracted Records to the Intermediate Files
  - Configuring Archived Log Switches
- Applier Processing of Intermediate Files
- Applier Handling of the Sync SCN Value
- Applier Processing of Updates in Merge Apply Mode
- Commit Processing and Target Constraints with Applier Multi-Threaded Processing
- Applier Processing of Updates for Duplicate Records
- Conflict Resolution for Replicated Data
- Post-Apply Processing
  - AFTER_APPLY XML Data Structure
  - Data Visibility Across Multiple Applier Threads
DDL Replication
- DDL Replication Overview
- Considerations for DDL Replication
- Replication Considerations for CREATE TABLE and ADD COLUMN Operations
  - Logging Supplemental Information for Source Tables Created with Replicated DDL
- Replication of TRUNCATE TABLE Operations
- Order in Which Data Replication Applies DDL Operations
- Examples of DDL Replication
- Supported DDL Operations for Sources
- Supported DDL Operations for Targets
Recovery and Checkpoint Processing
- Stages of Recovery Processing
- Checkpoint Processing
- Recovery Tables
- Guidelines for Resuming Applier Processing After a Failure
Configuring the Server Manager
- Server Manager Main Server and Subservers
- Installing the Server Manager as a Service on Windows
- Starting the Server Manager as a Service or Daemon
- Manually Starting the Server Manager
- Stopping a Server Manager Service or Daemon
- Uninstalling the Server Manager Service on Windows
- Defining the Main Server and Its Subservers
- Editing the Main Server or a Subserver
- Editing Microsoft SQL Server Instance Settings
- Editing Properties for the Main Server or a Subserver
- Viewing Information About the Server Manager System
- Configuring the Server Manager for HTTPS Communication
- Configuring the Server Manager Main Server to Run with NAT
- Creating and Managing User Accounts
- Associating a Subserver with Another Main Server
- Deleting Subservers
Creating Replication Configurations in the Data Replication Console
- Data Replication Console Interface
- Starting the Informatica Data Replication Console
- Task Flow: Creating a Replication Configuration
- Defining the Source Database
  - Configuring a Connection to an Oracle ASM Instance
  - Connecting to Oracle RAC Sources
    - Connecting to an Oracle RAC by Using Custom Connection Strings
    - Connecting to an Oracle RAC by Using a Virtual IP Address
- Defining the Target Database
- Generating Target Tables and Audit Log Tables
- Handling Source Tables with Long Table or Column Names in Merge Apply or Audit Apply Mode
  - Strategies for Handling Long Table Names
    - Editing the Audit Log Table Suffix
    - Manually Editing Target Table Names in an SQL Script for Merge Apply
  - Strategies for Handling Long Column Names
- Mapping Source and Target Tables
- Defining Source Table Indexes
- Customizing Apply Settings for Target Tables
- Configuring Conflict Resolution
  - Example of Configuring a MAXIMUM Resolution Strategy for Update Conflicts
  - Example of Configuring Custom Conflict Resolution
- Configuring the Start Points for Extractor and Applier Tasks
- Customizing Column Mappings
  - Filtering Column Data
- Enabling Replication of DDL Changes
- Adding Tcl and SQL Expressions
- Specifying the Database Logs from Which to Extract Data
- Configuring Runtime Settings
- Configuring Message Logging
- Saving Replication Configurations to the Main Server Manager
Managing Replication Configurations
- Configuration Management Tasks
- Opening a Replication Configuration in the Data Replication Console
- Switching to Read or Edit Mode for a Replication Configuration
- Editing a Replication Configuration
- Updating Connection Information for a Replication Configuration
- Clearing User Replication Settings
- Cleaning Replication-Related Information
- Deploying Replication Configurations
- Importing a Configuration File
- Exporting a Configuration File
- Replication Configuration Revisions
  - Viewing Earlier Revisions of a Replication Configuration
- Generating a Reverse-Replication Configuration
- Managing Database Supplemental Logging
Running Data Replications
- Methods of Running Replications
- Conflicting Replication Tasks
- Running Replication Executables Manually from the Data Replication Console
- Scheduling Replication Tasks
Configuring Selected Replication Types
- Overview of Selected Replication Types
- Configuring Continuous Replication
- Configuring Data Replication from One Source to Multiple Targets
- Loopback Avoidance for Replicated Data
- Configuring Bidirectional Replication
- Configuring Cascade Replication
Monitoring Data Replication
- Types of Monitoring Information
- Replication Statistics
- Intermediate Files
- Task Execution Logs
- Server Manager Logs
  - Viewing Server Manager Logs
- User Notifications
- Skipped Transaction Records
- Managing Open Transactions
Maintaining the Replication Environment
- Manually Changing the Source Table Structure After Running Data Replication
- Adding Table Mappings Manually After Running Data Replication
- Viewing a List of Processed Database Logs
- Resuming Replication After Upgrading a DB2 for Linux, UNIX, and Windows Source Database
Troubleshooting
Data Replication Files and Subdirectories
- Data Replication Script Files
- Executables Called from the Data Replication Console or Scripts
- Other Key Files
- Subdirectories
Data Replication Runtime Parameters
InitialSync, Extractor, and Applier Command Line Parameters
Server Manager Command Line Parameters
Updating Configurations in the Replication Configuration CLI
- Replication Configuration CLI Overview
- Updating Source and Target Metadata for a Replication Configuration in the CLI
  - Updating Source and Target Metadata for a Configuration in Interactive Mode
  - Updating Source and Target Metadata for a Configuration in Non-interactive Mode
- Replication Configuration CLI Commands
DDL Statements for Manually Creating Recovery Tables
Sample Scripts for Enabling or Disabling SQL Server Change Data Capture
- Microsoft SQL Server Enterprise Edition
Glossary
- Applier
- apply cycle
- Audit Apply
- audit log table
- bidirectional replication
- calculated columns
- cascade replication
- change data capture
- checkpoint processing
- Command Line Interface
- configuration file
- continuous replication
- data files
- data warehouse appliance
- Edit mode
- Extractor
- Flat File target
- global transaction
- heterogeneous replication
- Informatica Data Replication Console
- initial materialization
- InitialSync
- intermediate files
- loopback avoidance
- Merge Apply
- primary target
- Read mode
- recovery table
- Replication Configuration Command Line Interface
- replication configuration file
- replication statistics
- replication tasks
- routing
- secondary target
- Send File task
- Server Manager
- Server Manager Command Line Interface
- Server Manager Main server
- Server Manager subserver
- SQL Apply
- SQL Script Engine
- staging table
- Start SCN
- subtask threads
- supplemental logging
- Sync SCN
- Tcl expression
- Tcl Script Engine
- transaction files
- transactional replication
- virtual column
- virtual index

User Guide

9.6.3 HotFix 1
- 9.7.0 HotFix 1
- 9.6.3

Back Next

Data Visibility Across Multiple Applier Threads

Applier thread 1 runs post-apply SQL statements or stored procedures before the main Applier thread commits the data across all of the Applier threads. Consequently, Applier thread 1 can process only the data that was committed during previous apply cycles and the change data that Applier thread 1 applied to the target during the current apply cycle. Applier thread 1 cannot process change data that other Applier threads applied to the target during the current apply cycle.

Some target databases provide mechanisms for reading uncommitted data that was loaded by multiple threads.

For Teradata targets, you can use the LOCK TABLE FOR ACCESS clause to ensure that the post-apply SQL statements or stored procedures process change data that other Applier threads applied to the target during the current apply cycle.

The LOCK TABLE clause is a part of the DML SQL command. It is not a separate statement.

The following SQL statement inserts the total number of records in the tgt_db1.tgt_tab target table and the current timestamp into the tgt_db1.stat table:

LOCK TABLE tgt_db1.tgt_tab FOR ACCESS
INSERT INTO tgt_db1.stat(COUNTER,LOAD_TIME)  
SELECT count(*), current_timestamp(0) from tgt_db1.tgt_tab;

The total number of records includes the change data that was loaded to this table by all of the Applier threads during the current apply cycle even though this data is not committed yet.

Rename Saved Search

Table of Contents

User Guide

User Guide

Data Visibility Across Multiple Applier Threads

Data Visibility Across Multiple Applier Threads