In Review

In this chapter we looked at replication in theory and introduced the three replication types available to SQL Server solutions: snapshot replication, transactional replication, and merge replication. We also investigated two additional forms of redundant and fail-over database solutions, log shipping and database mirroring.

Microsoft has gone to great lengths to build a database management system that is ideal for distributed database solutions and wide-area deployment on networks like the Internet. While its replication architecture is not new or revolutionary, it goes without saying that without replication, log shipping, and mirroring, none of the other features we discuss later in the book would be worth exploring for Internet applications. It is very unlikely today that anyone deploying SQL Server on the Internet or on a WAN would not be configuring for replication and fail-over, especially if the business is in the hurricane corridor in the southeast, tornado alley in the midwest, or the earthquake environs out in the west. Out of all the subjects a SQL Server DBA might have to master, none other might be as challenging and rewarding as setting up a distributed fully redundant network. To complete the fail-over scenario, let's now look at clustering.