Learning MySQL, 2nd Edition

Chapter 13. High Availability

In the IT context, the term high availability defines a state of continuous operation for a specified length of time. The goal is not eliminating the risk of failure—that would be impossible. Rather, we are trying to guarantee that in a failure situation, the system remains available so that operation can continue. We often measure availability against a 100% operational or never-fails standard. A common standard of availability is known as five 9s, or 99.999% availability. Two 9s would be a system that guarantees 99% availability, allowing up to 1% downtime. Over the course of a year, this would translate to 3.65 days of unavailability.

Reliability engineering uses three principles of systems design to help achieve high availability: elimination of single points of failure (SPOFs), reliable crossover or failover points, and failure detection capabilities (including monitoring, discussed in Chapter 12).

Redundancy is required for many components to achieve high availability. A simple example is an airplane with two engines. If one engine fails while flying, the aircraft can still land at an airport. A more complex example is a nuclear power plant, where there are numerous redundant protocols and components to avoid catastrophic failures. Similarly, to achieve high availability of a database we need network redundancy, disk redundancy, different power supplies, multiple application and database servers, and much more.

This chapter will focus on the ...