What is an OLAP Cube?
When working with RevoGrid Pivot, you will often hear the term OLAP (Online Analytical Processing) or OLAP Cube. While the name sounds complex, the concept is fundamental to modern business intelligence.
The “Cube” Metaphor
Section titled “The “Cube” Metaphor”In a traditional spreadsheet, data is 2-dimensional (Rows and Columns). In an OLAP Cube, data is Multidimensional.
Think of a “Cube” where each side represents a different category (a Dimension):
- Product (what we sold)
- Time (when we sold it)
- Geography (where we sold it)
Inside the cube, at the intersection of these three dimensions, lies the Measure (e.g., Sales Amount).
By “rotating” this cube, you can view the data from any angle. In RevoGrid Pivot, this “rotation” happens when you move fields between the Rows and Columns axes.
Core OLAP Operations
Section titled “Core OLAP Operations”OLAP concepts
Slice, dice, drill and rotate the same analytical data
OLAP cubes are designed to support four primary analytical operations, all of which are supported by RevoGrid:
1. Slicing
Section titled “1. Slicing”Slicing is like taking a single “slice” of the cube.
- Concept: Picking one value for a dimension.
- Example: “Show me sales only for the Year 2024.”
- In RevoGrid: This is achieved through Filtering.
2. Dicing
Section titled “2. Dicing”Dicing is like cutting a smaller “sub-cube” out of the big one.
- Concept: Picking specific values across multiple dimensions.
- Example: “Show me sales for Electronics in the North Region during Q1.”
- In RevoGrid: This is achieved by combining filters and row/column grouping.
3. Drill-Down / Drill-Up
Section titled “3. Drill-Down / Drill-Up”Moving through the levels of a hierarchy.
- Concept: Going from summarized data to detailed data (Drill-Down) or back (Drill-Up).
- Example: Looking at sales by Year, then clicking to see Quarters, then Months.
- In RevoGrid: Enabled via the
collapsedproperty in Drill-Down.
4. Pivoting (Rotating)
Section titled “4. Pivoting (Rotating)”Changing the orientation of the axes.
- Concept: Swapping Rows and Columns to see the data differently.
- Example: Moving “Region” from the Row axis to the Column axis to compare regions side-by-side.
- In RevoGrid: Performed via the Configurator or by updating the
rowsandcolumnsarrays in your code.
Why “Cube” vs. “Flat Table”?
Section titled “Why “Cube” vs. “Flat Table”?”A cube is a way to describe analytical questions across several fields. It does not require a particular database, and it does not guarantee that results are precomputed.
| View | What it contains | Example |
|---|---|---|
| Source records | Individual events | Every order with Region, Product and Sales |
| Pivot report | Grouped summaries | Total Sales for each Region and Quarter |
| Analytical backend | A service that calculates the requested summaries | Your authorized SQL or OLAP endpoint |
SQL engines can calculate grouped summaries directly. Dedicated analytical engines may add precomputation or specialized storage. Performance depends on the data, query, indexes and backend design; the label “cube” alone does not make a query faster.
How RevoGrid Fits In
Section titled “How RevoGrid Fits In”RevoGrid displays and lets users explore the analytical report.
- Client-side: Pivot groups an array of source records in the browser.
- Server-side: Pivot sends the report request to your application API. Your backend authorizes the request and calculates the summaries using its chosen engine.
Use the server-side introduction for a first connection, then the remote contract for the request and response. An adapter connects your service to that contract; naming a database is not an out-of-the-box integration.
Next Steps
Section titled “Next Steps”Return to Prepare Your Data for local reports, or continue with server-side setup when calculation belongs on your backend.