Go to documentation repository
Replication is the process of data coping among databases that are in the same or different servers and connected via LAN, WAN or Internet. The following terms are used in MS SQL Server replication:
Term | Description |
Publication | Server (database) that sends data to another server (database) |
Distribution | Server that controls data flow via replication system. Contains a specialized distributor database. Distribution and Publication can coincide |
Subscription | Server (database) that receives data from the other server (database) |
Publication | Sum of articles, sent to Subscription |
Article | Main replication module: table or table's subset |
Subscription | Data group, received by Subscription |
Distribution database | System database that is saved at Distribution and doesn't contain user tables. Is used for saving the task snapshots and all the transactions, waiting for distribution by the Subscription |
Note.
On default Replication service is not installed together with MS SQL Server Express.
Note.
MS SQL Server Express cannot be used as Publication or Distribution.
In the Auto PSIM distributed architecture it is reasonable to use replication with topology «Central subscription». In this case several Publication servers (for example servers, installed on traffic control posts) copy the data to the central server – Subscription.
As it is necessary to transmit data among servers of databases, it is recommended to use transactions replication. This replication type is also preferable in case when copied data constantly change, size of copied data is sufficiently great and not necessary to support auto changes of replicated data regarding Publication and Subscription servers. All described terms are executed in the Auto PSIM distributed architecture.
The following agents are used while replicating: