UPDATE CUBE (MDX)

Syntax

UPDATE [ CUBE ] Cube_Name SET 
      <update clause> 
         [, <update clause> ...n ]
 
<update clause> ::= 
      Tuple_Expression[.VALUE]= New_Value
      [ 
        NO_ALLOCATION
      | USE_EQUAL_ALLOCATION 
      | USE_EQUAL_INCREMENT 
      | USE_WEIGHTED_ALLOCATION [ BY Weight_Expression] 
      | USE_WEIGHTED_INCREMENT [ BY Weight_Expression]
      ]

Arguments

• Cube_Name
  A valid string that provides the name of a cube.

• Tuple_Expression
  A valid Multidimensional Expressions (MDX) expression that returns a tuple.

• New_Value
  A valid numeric expression.

• Weight_Expression
  A valid Multidimensional Expressions (MDX) numeric expression that returns a decimal value between 0 and 1.

Remarks

The cell specified by the tuple expression can be any valid cell in the multidimensional space (that is, the cell does not have to be a leaf cell). However, the cell must be aggregated with the Sum aggregate function and must not include a calculated member in the tuple that is used to identify the cell.

It may be helpful to think of the UPDATE CUBE statement as a subroutine that will automatically generate a series of individual cell writeback operations to leaf and non-leaf cells that will roll up into a specified sum.

The following table describes the methods of allocation.

<leaf cell value> = 
<New Value> / Count(leaf cells that are contained in <tuple>)

<leaf cell value> = <leaf cell value> + 
(<New Value > - <existing value>) /
Count(leaf cells contained in <tuple>)

<leaf cell value> = < New Value> * Weight_Expression

<leaf cell value> = <leaf cell value> + 
(<New Value> - <existing value>)  * Weight_Expression

If a weight expression is not specified, the UPDATE CUBE statement implicitly uses the following expression:

Weight_Expression = <leaf cell value> / <existing value>

A weight expression should be expressed as a decimal value between zero (0) and 1. This value specifies the ratio of the allocated value that you want to assign to the leaf cells that are affected by the allocation. The client application programmer's has the responsibility of creating expressions whose rollup aggregate values will equal the allocated value of the expression.

Each UPDATE CUBE allocation should be considered to be atomic for transactional purposes. This means, that if any one of the allocation operations fails for any reason, such as an error in a formula or a security violation, the whole UPDATE CUBE operation will fail. Before the calculations of the individual allocation operations are processed, a snapshot of the data is taken to ensure that the resulting calculations are correct.

See Also

Reference

Other Resources