Cube's strength lies in its unique data model. By mirroring data from your source systems and embedding it into your customized data hierarchy, Cube seamlessly transforms this data into an OLAP cube in the cloud. The result? Instant access to crucial data, propelling informed business decisions with a simple click.
To optimize Cube for your needs, there are two pivotal decisions your team will make during the initial setup:
- Determining your data's primary categories or top-level dimensions.
- Designing the hierarchy outlining how data will be structured within each top-level dimension.
What is an OLAP cube?
OLAP is short for Online Analytical Processing, and an OLAP cube is a type of data structure that catalogs data at the intersections of dimensions or categories. Although "cube" suggests a specific shape, the form it takes adapts based on the number of dimensions.
Traditional spreadsheets offer a two-dimensional data representation, where data points correspond to row and column intersections.
An OLAP cube with three dimensions, such as accounts, departments, and time, will be shaped like a cube.
As dimensions increase, so does the OLAP cube's complexity. No matter how complex an OLAP cube becomes, the relational model continues to use the intersection of dimensions to locate values. The image below represents the relational model of a five-dimensional OLAP cube.
The dimensions we use to build your OLAP cube are called top-level dimensions because they sit at the top level of your data. During implementation, your team will determine up to eight top-level dimensions to serve as the foundation of your Cube.
If you need more than eight dimensions, you can purchase an additional Cube, and if you need to make changes to your dimensions after implementation, please reach out to your CSM.
Once top-level dimensions are in place, the associated data for each dimension needs to be organized. A coherent and well-maintained data hierarchy, or structure, expedites data retrieval.
You are probably already familiar with your source system's hierarchy and can choose to use a similar hierarchy in Cube. The Account top-level dimension commonly follows your Chart of Accounts, while the Time top-level dimension naturally orders chronologically. However, any top-level dimension can have a custom hierarchy.
If you are new to creating a data structure, you can visualize a warehouse. The most efficient way of locating a specific item in the warehouse requires knowing the item's aisle, shelf, and bin numbers. Similarly, navigating through an OLAP cube uses the top-level dimensions and subcategories to locate values.
Top-level dimension sub-categories
Each top-level dimension in the OLAP cube has subcategories called parent or or child dimensions. Child dimensions store values mapped from your source system or calculated using formulas. Parent dimensions have children nested below them, increasing the granularity of the data at that location.
Unlike child dimensions, parent dimensions don't house values directly. Instead, they derive these values from their child dimensions, including any dimensions calculated using a formula, ensuring up-to-date accuracy. Values undergo a cumulative rollup in hierarchies with multiple nested dimensions, aggregating data until all computations are completed.
The hierarchy of each top-level dimension is flexible. Using the Cube web portal, you can introduce or modify parent/child dimensions, rearrange them, or modify the linkage between them and your source data.
Learn more about how to use the Web portal to manage your dimensions.