Question

我有一个索引视图：

CREATE VIEW ptv.vw_mokiniai_2
 WITH SCHEMABINDING
 AS 
  SELECT 
    T1.year_name,
    T2.person_id,
    T2.year,
    T2.person_position,
    COUNT_BIG(*) AS [aggregated number]
  FROM Table1 AS T1 
  INNER JOIN Table2 AS T2
     ON T1.year = T2.year
  GROUP BY T1.year_name,
    T2.person_id,
    T2.year,
    T2.person_position

列上有唯一的聚簇索引

T1.year_name, T2.person_id, T2.year, T2.person_position

我需要查询优化器在查询

上使用此索引视图

SELECT 
    T1.year_name, 
    COUNT_BIG(DISTINCT T3.person_id) AS [persons_per_year]
FROM Table1 AS T1
INNER JOIN (
            SELECT 
              T2.person_id,
              T2.year,
              T2.person_position
            FROM Table2 AS T2 
            GROUP BY T2.person_id, T2.year, T2.person_position
            ) AS T3 
    ON T1.year = T3.year
GROUP BY T1.year_name

查询必须是派生表，因为COGNOS生成此查询，我不想编辑多维模型。有人知道怎么办？也许优化查询的唯一方法是删除派生表（意味着我必须编辑多维模型）？任何建议（视图编辑和最终查询编辑）将不胜感激。

Answer 1

以下示例使用我的测试中的索引视图。如果这在您的环境中不起作用，那么您发布的代码中可能存在差异。

CREATE TABLE dbo.Table1
    (
      year varchar(10)
    , year_name varchar(10)
    );

CREATE TABLE dbo.Table2
    (
      year varchar(10)
    , person_id varchar(10)
    , person_position varchar(10)
    );
GO

CREATE VIEW dbo.vw_mokiniai_2
WITH SCHEMABINDING
AS
    SELECT  T2.person_id
          , T2.year
          , T2.person_position
          , COUNT_BIG(*) AS [aggregated number]
    FROM    dbo.Table2 AS T2
    GROUP BY T2.person_id
          , T2.year
          , T2.person_position
GO
CREATE UNIQUE CLUSTERED INDEX cdx ON vw_mokiniai_2(year, person_id, person_position)
GO

SELECT 
    T1.year_name, 
    COUNT_BIG(DISTINCT T3.person_id) AS [persons_per_year]
FROM Table1 AS T1
INNER JOIN (
            SELECT 
              T2.person_id,
              T2.year,
              T2.person_position
            FROM Table2 AS T2 
            GROUP BY T2.person_id, T2.year, T2.person_position
            ) AS T3 
    ON T1.year = T3.year
GROUP BY T1.year_name
GO

索引视图可用性

1 个答案: