In Review

This chapter deals with the high-availability features now available in SQL Server 2005. We discussed the various scale-up and scale-out options and investigated the myth of linear scale-up and how adding more and more processors to a system has a point of diminishing return, both in terms of price and performance.

We also investigated the various clustering models possible using the cluster services of either Windows Server 2003 Enterprise Server or Datacenter Server. Creating virtual SQL Server servers and setting up the cluster server has been getting easier with each version of SQL Server since SQL Server 7.0.

If you do not need a cluster solution, or have budget constraints, you can configure standby (warm) servers that can be configured to replace the primary server if it goes down. Availability is not as immediate and there will also be delays and additional work involved in backing up databases and transaction logs and applying them to the standby server. The standby server will also have to be kept in read-only or standby mode so that the data cannot be modified, rendering it out of sync with the primary server.

This chapter caps our sojourn into the life of the SQL Server database administrator and his or her typical duties. The next parts deal with programming.