Synthetic Test Data is an object-oriented data analysis framework that can be used to generate test data by using the MDD (model, design, data) approach. It provides an easy interface for both designers and testers to work on the same project. The key feature of Synthetic Test Data is its usability and scalability with a small footprint. The following discussion highlights the characteristics of Synthetic Data and some advantages of using it.
Designing test cases with MDD can become time consuming and tedious. Synthetic Data alleviates this problem by providing a simplified tool for designing test cases with defined parameters. Also, you do not need to concern yourself with complex data extraction methods such as Regression, Latent Semantic Analysis, correlated variables, or even normalization. Synthetic Data has a number of easy-to-use convenience functions that help you with data extraction conveniently.
Synthetic Data comes with easy and convenient functionality for managing test cases, data sources, and test execution environment. It helps you manage test data efficiently and minimizes the risks of data corruption due to unexpected changes in the input data. Also, it simplifies the process of data extraction, data manipulation, and data validation. You also get an easy to use wrapper API for creating custom objects.
In addition, Synthetic Data has extensive support for various databases including Microsoft SQL Server. With the help of this framework, you can easily construct, manipulate, and combine various types of data into one logical database. In addition, the framework provides support for other data types such as binary/hierarchical, text, ternary, binary/ hexagonal, and numeric data. Synthetic Data includes many utility classes for constructing, managing, and maintaining database driven applications. These classes help you with common database related issues such as planning and construction, managing transaction logs, and implementing transactions.
The main goal of Synthetic Data is to provide an easy to use, adaptable, reusable, dynamic way of testing real world data sets. In fact, Synthetic Data is a very powerful part of the MCTPE (Multi-Stage Data Integration) system. This system includes a set of generic commands that can be used to transform any data into a format of your choice. Furthermore, the framework supports a wide variety of transformations. It makes it very easy to create, modify, and extend the language.
Another major advantage of using Synthetic Data is that it provides quick implementation of complex business requirements. It provides data extraction, validation, transformations, and data loading that are usually complex and time consuming to perform on your own. Synthetic data is also designed so that you can easily define test cases and make them run quickly. Also, the framework supports data extraction over a large number of levels in both RDBMS and text formats. In fact, it can handle all kinds of data structures including XML documents, association tables, and complex data structures.
The MCTPE framework supports integration of any source of data into your test cases. However, this feature can be useful only when you have multiple sources of data and it becomes tedious to maintain them in your test code. You can easily define sources of data in the Synthetic Data editor. Additionally, the Synthetic Data editor allows you to conveniently change the format of the underlying data without modifying your code or the Synthetic Data itself.
The MCTPE framework supports data extraction and transformations that work fine for simple small-to-medium sized test cases. However, if you need to test very complex business logic, the MCTPE will fail to give you consistent results. For example, it will not be possible to test for over 50 relationships at once because it requires too many data sets to run. However, the lack of options makes it suitable for integration testing only.