![]() Verify that data is transformed correctly according to system requirements and business rules.Ensure that the ETL application properly rejects, replaces with default values and reports invalid data.Compare record counts between data sources.Are the requirements clear (is there any ambiguity)?. ![]() Some levels of testing are outlined below, but this should not be considered an exhaustive list of testing activities. There are several levels of testing that should be performed during ETL testing. There are a variety of tools that can be used for ETL testing, but that decision can be left to the organization and the test team. Testing, preferably by an independent party, should be undertaken to verify and validate the ETL process, thereby ensuring the quality, completeness and robustness of the data warehouse. In the data warehouse model, the cost related to the failure of discovering a defect early in the process is often worsened by the fact that key business decisions are potentially being made based on the data presented. In software development, the cost of finding and fixing a defect rises exponentially as the development lifecycle progresses. Because data is coming from various sources, and in various forms, it is vital that the ETL process is working correctly. For those who are unfamiliar with the term, ETL stands for Extract-Transform-Load, and is the process for consolidating all of the data from various sources into the data warehouse model. This leads us to the importance of ETL testing. It usually contains historical data derived from transaction data, and enables an organization to consolidate data from several sources. The data warehouse is designed for query and analysis rather than for transaction processing. This model is referred to as the data warehouse (we will not get into the complexities of ETL/data warehouse architecture). To be utilized as such, data must be gathered from a number of different sources, then transformed and stored in a single database so that the organization can monitor, analyze and report on the data. It is common now for an organization to collect and analyze real-time data for corporate decision making, reporting, data mining and for reviewing historical trends. ![]() A database is one of the most important assets an organization may own, and the data contained within that database is generally invaluable. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |