It is useful to divide replication into two broad categories: replicating data in a server to server environment and replicating data between a server and clients. This section of the documentation describes scenarios that involve replicating data between servers. Data is typically replicated between servers to support the following applications and requirements:
Improving scalability and availability
Maintaining continuously-updated copies of data allows read activity to be scaled across multiple servers. The redundancy resulting from maintaining multiple copies of the same data is crucial during planned and unplanned system maintenance. For more information, see Improving Scalability and Availability.
Data warehousing and reporting
Data warehouse and reporting servers often use data from online transaction processing (OLTP) servers. Use replication to move data between OLTP servers and reporting and decision support systems. For more information, see Data Warehousing and Reporting.
Integrating data from multiple sites
Data is often "rolled up" from remote offices and consolidated at a central office. Similarly, data can be replicated out to remote offices. For more information, see Integrating Data from Multiple Sites (Server).
Integrating heterogeneous data
Some applications depend on data being sent to or from databases other than SQL Server. Use replication to integrate data from non-SQL Server databases. For more information, see Integrating Heterogeneous Data.
Offloading batch processing
Batch operations are often too resource intensive to run on an OLTP server. Use replication to offload processing to a dedicated batch processing server. For more information, see Offloading Batch Processing.
For information on server to client scenarios, see Replicating Data Between a Server and Clients.