Validation and system testing:
Validation testing is a concern which overlaps with integration testing. Ensuring that the application fulfils its specification is a major criterion for the construction of an integration test. Validation testing also overlaps to a large extent with system testing, where the application is tested with respect to its typical working environment. Consequently for many processes no clear division between validation and system testing can be made. Specific tests which can be performed in either or both stages include the following.
* Regression testing. Where this version of the software is tested with the automated test harnesses used with previous versions to ensure that the required features of the previous version are still working in the new version.
* Recovery testing. Where the software is deliberately interrupted in a number of ways, for example taking its hard disc off line or even turning the computer off, to ensure that the appropriate techniques for restoring any lost data will function.
* Security testing. Where unauthorised attempts to operate the software, or parts of it, are attempted. It might also include attempts to obtain access the data, or harm the software installation or even the system software. As with all types of security it is recognised that someone sufficiently determined will be able to obtain unauthorised access and the best that can be achieved is to make this process as difficult as possible.
* Stress testing. Where abnormal demands are made upon the software by increasing the rate at which it is asked to accept data, or the rate at which it is asked to produce information. More complex tests may attempt to create very large data sets or cause the software to make excessive demands on the operating system.
* Performance testing. Where the performance requirements, if any, are checked. These may include the size of the software when installed, the amount of main memory and/ or secondary storage it requires and the demands made of the operating system when running within normal limits or the response time.
* Usability testing. The process of usability measurement was introduced in the previous chapter. Even if usability prototypes have been tested whilst the application was constructed, a validation test of the finished product will always be required.
* Alpha and beta testing. This is where the software is released to the actual end users. An initial release, the alpha release, might be made to selected users who would be expected to report bugs and other detailed observations back to the production team. Once the application has passed through the alpha phase a beta release, possibly incorporating changes necessitated by the alpha phase, can be made to a larger more representative set users, before the final release is made to all users.
The final process should be a software audit where the complete software project is checked to ensure that it meets production management requirements. This ensures that all required documentation has been produced, is in the correct format and is of acceptable quality. The purpose of this review is: firstly to assure the quality of the production process and by implication the product; and secondly to ensure that all is in order before the initial project construction phase concludes and the maintenance phase commences. A formal hand over from the development team at the end of the audit will mark the transition between the two phases.
|