Application test data creation and data masking

Managing test data in software applications is one of the important subtitles of the testing process. The preparation of test data requires a serious effort and therefore a great amount of time can be lost in test processes. There are points to be considered in the creation and transfer of test data. The use of personal data through information systems is very common, regardless of the public and private sector. Although there are some advantages and conveniences provided by data usage in this way, this may increase the risk of data security. Many organizations prefer to use data masking to minimize data security costs and protect their data. In this thesis, it is aimed to create the data in the live environment database in the test environment database by masking it with appropriate data masking method and masking techniques. In addition, an exemplary comparison has been made on how the performance can be increased in the masking process, that is, how the time cost can be reduced. As a material in the study, a sample data set created similar to a living environment database was used. Data masking method and techniques have been determined in order to use the data masking application determined in accordance with the study content on the sample data set. The selected data masking application was applied to the data set created with the static data masking method and masking techniques, and then the masked data was transferred to the test environment database. Applied data masking techniques were compared under separate headings, as sample data set and masked data set. The masking and the transfer of masked data were first performed with multiple threads, independent and parallel operation, then this process was repeated with a single thread without parallel operation. It has been observed that data masking and transfer of masked data performed in parallel with multiple threads increase the temporal performance by 32% compared to the transfer process of masking and masked data without parallel operation with a single thread, in other words, it saves time by 32%.