Training Database

During the design phase of the development, the IT department decided they would use a Microsoft Access platform to create the training database. The decision to apply this platform was base on the extensive knowledge that the technicians already had and the established success of the Access. Having a proven program would reduce some of the testing but it would still require and extensive trials with our training database. This was especially necessary, in the functions that interfaced with the internet and other Air Force databases.

When developing the system, the IT department took a few different approaches to testing. The initial approach was unit testing. Unit testing is defined as the testing of individual code modules or methods before they are integrated with other modules (Satzinger, 2004). This is exactly how the IT department explained their testing to me. As each module of the database was created, it was then tested for functionality and error.

Upon successful unit testing, the individual modules were then tested for compatibility and continuity with other modules. This is what the text refers to as integration testing. Although the modules tested okay separately, they did not always test properly with the other modules. Once interfaced with other functions, the possibility for error is dramatically increased. Some of the errors could include but are not limited to interface incompatibility, invalid parameter values, runtime errors or unexpected state interactions (Satzinger, 2004). Once this portion of the testing was completed the designers could then move on to testing the system as a whole.

Since the size of the project was relatively small and it would operate mostly in-house, the IT department decided to skip further testing and release a Beta version of the database. At the time the database was not mission critical and system failure or crashing would not affect the unit's day-to-day operations. It was decided that the beta would serve as the testing tool to work out any remaining errors. A controlled number of users were giving access to the database and had the program installed on their terminals. For 30 days the IT departments trained the controlled group on the features and operation of the training database. The initial group of users was not only being trained on how to use the database; they were also being trained on how to train the database. The Beta version not only served as a testing tool, it also served as a valuable training tool.

After the majority of the faults were worked out of the system, it was then time to move to the installation phase. The Installation approach that ours most closely resembled was direct installation. Since we were not replacing another automated system we did not have any data to transfer from another database. The only data that was going to be imputed