Features and Examples of Using Various Summation Types

Remember that element summation is unavailable for some source cube dimensions.

To execute summation, the following conditions must be satisfied:

If any of these conditions is not satisfied, calculation is not executed.

NOTE. It is recommended to use recursive dimensions.

See below examples of various summation methods in formulas of multidimensional calculation on database server for cubes having one or two recursive dimensions.

Summation for Cube with One Recursive Dimension

Multidimensional calculation uses the same cube as the source one and the destination one. This cube consists of a fact and a recursive dimension:

Before calculation the cube contained the following data:

The following formula is defined in the multidimensional calculation formula editor:

Destination cube coordinate Specified formula
b1 b1

where b1 is an element selected in dimensions. The expression is defined based on the existing cube fact.

The following are various types of summation using the same input data and the defined formula.

The calculated value is underlined in red. As you can see, the result uses the value of the b1 dimension element, which is a parent.

The calculated value is underlined in red. As you can see, the result uses the value of the b1 dimension element, which is a parent.

The calculated value is underlined in red. To understand the calculation, look at the dimension element identifiers. The b1 element has the 2 identifier, therefore the elements with the identifiers starting with 2 were used (these are b2 and b4), the value of the element defined in the formula was also used for calculation (this was b1). This implies: 100 + 10 + 30 = 140.

NOTE. Remember that identifiers are checked for all dimension elements.

Summation for a Cube with Two Recursive Dimensions

Multidimensional calculation uses the same cube as the source one and the destination one. This cube consists of a fact and two recursive dimensions (dimension1 is on the left, dimension2 is on the right):

Before calculation the cube contained the following data:

The following formula is defined in the multidimensional calculation formula editor:

Destination cube coordinate Specified formula
b1|a1 b1|a1

where b1 and a1 are elements selected in dimensions. The expression is defined based on the existing cube fact.

The following are various types of summation using the same input data and the defined formula.

The calculated value is underlined in red. The parent elements are located in the B1:B4 and C1:E1 cells, and the child elements are located in the C2:E4 range. As you can see, the result includes all parent and child values.

The calculated value is underlined in red. The parent elements are located in the B1:B4 and C1:E1 cells, and the child elements are located in the C2:E4 range. As you can see, the parent values have not been used to calculate the result.

The calculated value is underlined in red. To understand the calculation, look at the dimension element identifiers. The b1 element has the 2 identifier, therefore elements with identifiers starting from 2 were used for the dimension1 (these are b2 and b4). The a1 element has the 1 identifier, therefore elements with identifiers starting from 1 were used for the dimension2 (these are a2 and a4). The result also includes the values of child elements, which are at the intersection of the parent elements meeting the condition (matched identifiers). The value of the element defined in the formula is used for calculation (b1|a1). This implies: 1000 + 100 + 300 + 10 + 30 + 1 + 1 + 1 + 1 = 1444. The values included in the result except for b1|a1 are underlined in green.

See also:

Editing Formulas