Question

我们正面临其中一个MDX查询的问题。如果我们在where子句的Range中使用日期选择器，如果我们使用逗号作为分隔符，则执行else会花费6秒，大约需要1秒。

是否有办法通过在where子句中给出范围来修改查询并提高性能。请告诉我们您的意见。

示例1使用逗号分隔符

WITH MEMBER Measures.X AS SUM({[User].[column].CURRENTMEMBER},[Measures].[# User Count]) 
    SELECT NON EMPTY 
   UNION({[User].[column].[ALL]}, 
   Generate(Hierarchize([User].[Column].[ALL]) 
   AS [XL_Filter_Set_0], TopCount(Except(DrilldownLevel([XL_Filter_Set_0].Current 
   AS [XL_Filter_HelperSet_0],,0,INCLUDE_CALC_MEMBERS), [XL_Filter_HelperSet_0]) 
   , 5,  [Measures].[# User Count])) 
   )DIMENSION PROPERTIES PARENT_UNIQUE_NAME,HIERARCHY_UNIQUE_NAME ON ROWS  , Measures.X  ON COLUMNS 
  FROM [Activity] where 
  ({[Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2012],[Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2013], 
  [Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2014]}, 
  {[Activity Type].[Activity Hierarchy].[All]})

示例2没有逗号分隔符，但我们给出了范围

WITH MEMBER Measures.X AS SUM({[User].[column].CURRENTMEMBER},[Measures].[# User Count]) 
    SELECT NON EMPTY 
      UNION({[User].[column].[ALL]}, 
         Generate(Hierarchize([User].[Column].[ALL]) 
            AS [XL_Filter_Set_0], TopCount(Except(DrilldownLevel([XL_Filter_Set_0].Current 
             AS [XL_Filter_HelperSet_0],,0,INCLUDE_CALC_MEMBERS), [XL_Filter_HelperSet_0]) 
            , 5,  [Measures].[# User Count])) 
              )DIMENSION PROPERTIES PARENT_UNIQUE_NAME,HIERARCHY_UNIQUE_NAME ON ROWS  , Measures.X  ON COLUMNS 
          FROM [Activity] where 
  ({[Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2012]:[Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2014]}, 
  {[Activity Type].[Activity Hierarchy].[All]})

此致南西

Answer 1

而不是在WHERE中使用范围尝试使用FROM

中的范围

from (
select ([Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2012]:[Date].[Fiscal Hierarchy].[Fiscal Year Name].&[2014]) on Columns From [Activity])

查询与MDX性能有关

1 个答案: