Designing the Data Warehouse in Timextender Classic¶

Once you have set up your data warehouses and business units and established connections to the sources you want to use, it is time to start designing your data warehouse. This involves transferring data from the source systems to the data warehouse(s) via a staging database or ODX and applying transformations and data cleansing rules along the way.

It is common to set up more than one Data Warehouse to accomplish all of the transformations needed to get your data into the single form of truth that will be the semantic layer.

The first data warehouse is normally referred to as the Data Staging Layer or DSA. This is the first layer where data from your data sources is first loaded, and where you want to make sure all of your primary keys, data types, and incremental loading rules as appropriate are configured.

The second data warehouse is normally referred to as the Modern Data Warehouse or MDW. It is in this layer where the more complex transformations, lookups, and inserts are implemented. Views can also be created to combine data into a single table. Object dependencies may also be needed to ensure objects are executed in the correct order.

Additional data warehouses can also be added as needed, i.e. SDW. For example, dragging a view from one data warehouse layer into a subsequent data warehouse will transform the view into a table. Using multiple data warehouse layers can help facilitate the combination and transformation of data from disparate sources into a relatively small number of tables.