A table replication definition describes the table and the columns that are to be replicated. A primary table is the replication source; a replicate table is the destination. You can create one or more replication definitions for each primary table.
Create a replication definition in the Replication Server that manages the database where the primary table is stored.
The replication definition includes:
A name for the replication definition
The names of the primary and replicate tables, if they are different from each other and from the replication definition name
The location of the primary table
The names and datatypes of the primary columns to be replicated and the corresponding replicate column names
The names of the columns that form the primary key for the table
The replication definition can optionally include:
The names of columns that can be referenced in where clauses for subscriptions
Whether the replication definition and its columns will be used in replicating to a standby database
Whether to replicate all columns or the minimum number of required columns for update and delete operations
Replication status for text, unitext, image, and rawobject columns
Whether to change the datatype of replicated values from the datatype of the primary database to the datatype of the replicate database.
No data is distributed when you create a replication definition. You must create a copy of the table in each replicate database and then create subscriptions to begin replicating data.
Use these commands to work with table replication definitions:
create replication definition – creates a replication definition for a table.
alter replication definition – changes a replication definition.
drop replication definition – removes a replication definition.
For commands that you use in subscribing to replication definitions, see “Subscription commands”.
