Each Scheduler job is set up to send data from one database to another. The job can also be set up to run the data between multiple database locations.
To run data between multiple locations you can use Distribution Groups and Distribution Subgroups. You can group distribution locations together in various ways, depending on the physical location of the stores, their size, and so on. This can be useful, if there are many locations in your business and you need to control the distribution of your data.
The Scheduler job is set up with one or more Scheduler subjobs. The subjob represents a table that should be replicated between the hosts.
Each Scheduler subjob can be run as Normal or as Action.
Normal subjobs
Normal subjobs take all the data in the source database table and insert the data into the destination database table.
The data are exchanged between tables by comparing two tables and making them identical. Since this method often requires reading all table records in both databases, it is not recommended for big tables. However, for tables that neither create Actions nor have a Replication Counter, and from which records may be deleted, normal replication is the only option.
The exchange is further specified in the What To Do field.
Note: The consequences of deleting can be disastrous. Although deleting is an essential tool and very helpful during implementation and upgrades, it requires a lot of knowledge about the distributed environment to be used safely.
Action subjobs
Action subjobs only run the new or modified data. To do that, the job uses preactions to see what data have been added or updated since the job was last run.
Using this method limits the number of records that need to be replicated, and the method is useful when replicating master data from head office to stores.