Improving Scalability and Availability
Many applications can benefit from increased 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 also useful during planned and unplanned outages. If one server is unavailable, requests can be routed to a different server with the same data until the original server (or a replacement) is back online.
There are a number of ways to improve scalability and availability using replication, depending on application requirements:
If your application has more read activity than write activity, data caching through replication to multiple servers can be used to improve application scalability. For more information, see Improving Scalability.
If your application requires data availability during planned or unplanned outages of the database, see Improving Availability.
If your application requires both higher read scalability and the ability to handle planned and unplanned outages, see Improving Both Availability and Scalability.