Guideline: In More Details - Availability Of Test Environments

When the availability of the test environment is specified in a service level, it is important to agree on how availability will be measured. In other words, how do both parties define the availability of the test environment? This might differ. For instance, one party might feel that having the most important components available is adequate, while the other feels everything should be available. These two different interpretations yield a different result in the figure 1 - Schematic overview of the availability of various components in a test environment. In this figure, a test environment consists of five components (A, B, C, D and E). A, B, and C are identified as the main components. These are e.g. components on which the correct operation of the test object depends directly. Examples are a database server, web service, or mailboxes. Components D and E are considered less important, e.g. interfaces with other systems.

Figure 1: Schematic overview of the availability of various components in a test environment.

The availability is measured for each of these components. For instance, C and E are available continuously, but some problems (interruption of the line) occur in the other three. When looking at all components taken together, the total of failures are bigger than when looking solely at the main components.

The difference in availability is important mainly for different test levels. For a system test, for instance, it is less important that the interfaces are available continuously than for an end-to-end test.