Update Cube.md

Cube.md

@@ -1,18 +1,86 @@
 # Keyword `CUBE`
 
-AlaSQL supports CUBE() grouping options:
+## Introduction
+The `CUBE` operation in AlaSQL is an extension of the `GROUP BY` clause that allows you to generate multiple grouping sets in a single query. It's particularly useful for generating subtotals and grand totals in multidimensional data analysis, such as in OLAP (Online Analytical Processing) applications.
+
+## Syntax
+```sql
+SELECT column1, column2, ..., aggregate_function(column)
+FROM table
+GROUP BY CUBE(column1, column2, ...)
+```
+
+## Explanation
+The `CUBE` operation generates all possible combinations of the specified dimensions. For `n` dimensions, it produces 2^n grouping sets. This includes:
+- The grand total
+- Subtotals for all combinations of dimensions
+- Individual totals for each dimension
+
+## Examples
+
+### Basic CUBE Operation
 ```js
 var testData = [
     { Phase: "Phase 1", Step: "Step 1", Task: "Task 1", Val: 5 },
     { Phase: "Phase 1", Step: "Step 2", Task: "Task 2", Val: 20 },
     { Phase: "Phase 2", Step: "Step 1", Task: "Task 1", Val: 25 },
     { Phase: "Phase 2", Step: "Step 2", Task: "Task 2", Val: 40 }
 ];
-
-res = alasql('SELECT Phase, Step, SUM(Val) AS Val FROM ? \
+var res = alasql('SELECT Phase, Step, SUM(Val) AS Val FROM ? \
                   GROUP BY CUBE(Phase,Step)', [testData]);
 ```
 
-Try this example [in jsFiddle](http://jsfiddle.net/agershun/1nccgs6n/2/)
+Result:
+```js
+[
+  { Phase: 'Phase 1', Step: 'Step 1', Val: 5 },
+  { Phase: 'Phase 1', Step: 'Step 2', Val: 20 },
+  { Phase: 'Phase 1', Step: null, Val: 25 },
+  { Phase: 'Phase 2', Step: 'Step 1', Val: 25 },
+  { Phase: 'Phase 2', Step: 'Step 2', Val: 40 },
+  { Phase: 'Phase 2', Step: null, Val: 65 },
+  { Phase: null, Step: 'Step 1', Val: 30 },
+  { Phase: null, Step: 'Step 2', Val: 60 },
+  { Phase: null, Step: null, Val: 90 }
+]
+```
+
+In this result:
+- Rows with both Phase and Step represent individual groupings
+- Rows with Phase but null Step represent subtotals for each Phase
+- Rows with Step but null Phase represent subtotals for each Step
+- The last row with both null represents the grand total
+
+### CUBE with More Dimensions
+```js
+var res = alasql('SELECT Phase, Step, Task, SUM(Val) AS Val FROM ? \
+                  GROUP BY CUBE(Phase, Step, Task)', [testData]);
+```
+
+This query will generate 2^3 = 8 grouping sets, including all possible combinations of Phase, Step, and Task.
+
+## Comparison with ROLLUP
+While `CUBE` generates all possible combinations of dimensions, `ROLLUP` generates a subset of those combinations. `ROLLUP` is hierarchical, moving from the most detailed level to the grand total.
+
+For example, `ROLLUP(Phase, Step)` would produce:
+- (Phase, Step)
+- (Phase)
+- ()
+
+Whereas `CUBE(Phase, Step)` produces:
+- (Phase, Step)
+- (Phase)
+- (Step)
+- ()
+
+## Performance Considerations
+- `CUBE` can generate a large number of rows, especially with many dimensions. Use it judiciously to avoid performance issues with large datasets.
+- Consider using `GROUPING SETS` for more control over which specific combinations are generated.
+
+## Try it out
+You can try the basic CUBE example [in jsFiddle](http://jsfiddle.net/agershun/1nccgs6n/2/).
 
-See also: [ROLLUP](Rollup), [GROUPING SETS](Grouping Sets), [GROUP BY](Group By)
+## See Also
+- [ROLLUP](Rollup)
+- [GROUPING SETS](Grouping Sets)
+- [GROUP BY](Group By)