13 Availability

 

This chapter covers

  • The difference between high availability and disaster recovery
  • Availability requirements
  • Testing disaster recovery
  • When AlwaysOn availability groups are appropriate
  • Overloading clusters

High availability (HA) and disaster recovery (DR) are key concepts that every database administrator (DBA) should fully understand to make the correct implementation choices and protect their environments from single points of failure. The topic is both broad and deep, and this chapter does not attempt to focus on the actual implementation of HA and DR. Instead, we will focus on some of the key mistakes that DBAs make when planning their HA and DR strategy. For guidance on how to implement HA and DR in SQL Server, I recommend my book SQL Server 2019 AlwaysOn, which can be found at https://mng.bz/znaQ.

As SQL Server has matured, the scope of concepts and technologies that DBAs need to understand to implement HA and DR has increased dramatically. For example, to implement AlwaysOn availability groups (which we will now refer to as just availability groups), DBAs need to have an understanding of Windows clustering. Even if the implementation is a collaboration with a Windows team, DBAs must still understand the concepts.

13.1 #89 Confusing HA and DR

13.2 #90 Failing to architect for the requirements

13.3 #91 Not testing the DR strategy

13.4 #92 Assuming availability groups are always the right answer

13.5 #93 Overloading a cluster

Summary