Question

我正在浏览一些没有注释的pl / sql代码。试图理解它并优化它。以下是样本：

INSERT INTO gtt1 --75711 rows
(USER_ID, role_id, participant_code, status_id )
SELECT
 r.user_id, r.role_id, r.participant_code, MAX(status_id)
FROM
  user_role r,
  cmp_role c
WHERE
  r.role_id = c.role_id
  AND r.participant_code IS NOT NULL
  AND c.group_id = 3
  GROUP BY
  r.user_id, r.role_id, r.participant_code;

然后

DELETE gtt1
WHERE ROWID IN (SELECT ROWID FROM gtt1
                MINUS
                SELECT a.ROWID FROM gtt1 a, UIV_CMP_USER_ROLE b
                WHERE a.status_id = b.status_id
                AND (b.ACTIVE = 1 OR ( b.ACTIVE IN ( 0,3 ) 
                      AND SYSDATE BETWEEN b.effective_from_date AND b.effective_to_date
                     )
                )
                );

最后（这需要最长时间）

OPEN cv_1 FOR

SELECT c.role_id,
       c.subgroup,
       c.subgroup_description,
       COUNT(a.USER_ID) user_count
FROM   
    (SELECT b.user_id, b.role_id FROM gtt1 b, pt_user e
    --pt_user table has 73000 rows
        WHERE  e.user_id = RTRIM(b.user_id)
       ) a
RIGHT OUTER JOIN CMP_ROLE c ON a.role_id = c.role_id
WHERE c.group_id = v_group_id
GROUP BY c.role_id,c.subgroup,c.subgroup_description
ORDER BY c.subgroup;

有没有办法可以避免从gtt1中删除，最初只是获取我们想要的行？

运行解释计划我注意到对此查询进行了一些全表扫描：

SELECT 
   r.user_id, r.role_id, r.participant_code, MAX(status_id) 
  FROM 
    user_role r, 
    cmp_role c 
  WHERE 
    r.role_id = c.role_id 
    AND r.participant_code IS NOT NULL 
    AND c.group_id = 3 
    GROUP BY 
    r.user_id, r.role_id, r.participant_code 
    HAVING MAX(status_id) IN (SELECT b.status_id FROM UIV_CMP_USER_ROLE b 
                              WHERE (b.ACTIVE = 1 OR ( b.ACTIVE IN ( 0,3 )  
                                     AND SYSDATE BETWEEN b.effective_from_date AND b.effective_to_date 
                                    )) 
                             )

user_role = 803507行

cmp_role = 27行

user_role有5个索引：

idx 1 = role_id

idx 2 = last_updt_user_id

idx 3 = actv_id，participant_code，effective_from_Date，effective_to_date

idx 4 = user_id，role_id，effective_from_Date，effective_to_date

idx 5 = participant_code，user_id，roke_id，actv_cd

Answer 1

您想要优化查询本身还是只需要给出结果所需的时间？其次需要一些关于行数的数据，等等......

Answer 2

INSERT和DELETE在我看来等同于：

INSERT INTO gtt1 --75711 rows
(USER_ID, role_id, participant_code, status_id )
SELECT
 r.user_id, r.role_id, r.participant_code, MAX(status_id)
FROM
  user_role r,
  cmp_role c
WHERE
  r.role_id = c.role_id
  AND r.participant_code IS NOT NULL
  AND c.group_id = 3
  GROUP BY
  r.user_id, r.role_id, r.participant_code
  HAVING MAX(status_id) IN (SELECT b.status_id FROM UIV_CMP_USER_ROLE b
                            WHERE (b.ACTIVE = 1 OR ( b.ACTIVE IN ( 0,3 ) 
                                   AND SYSDATE BETWEEN b.effective_from_date
                                               AND b.effective_to_date
                                  ))
                           );

这是否更有效率我不知道 - 我需要更多地了解表格，索引和数据。

然后你可以进一步将GTT变成子查询，例如：

WITH gtt1 AS 
  (SELECT
   r.user_id, r.role_id, r.participant_code, MAX(status_id)
  FROM
    user_role r,
    cmp_role c
  WHERE
    r.role_id = c.role_id
    AND r.participant_code IS NOT NULL
    AND c.group_id = 3
    GROUP BY
    r.user_id, r.role_id, r.participant_code
    HAVING MAX(status_id) IN (SELECT b.status_id FROM UIV_CMP_USER_ROLE b
                              WHERE (b.ACTIVE = 1 OR ( b.ACTIVE IN ( 0,3 ) 
                                     AND SYSDATE BETWEEN b.effective_from_date AND b.effective_to_date
                                    ))
                             )
  )
SELECT c.role_id,
       c.subgroup,
       c.subgroup_description,
       COUNT(a.USER_ID) user_count
FROM   
    (SELECT b.user_id, b.role_id FROM gtt1 b, pt_user e
    --pt_user table has 73000 rows
        WHERE  e.user_id = RTRIM(b.user_id)
       ) a
RIGHT OUTER JOIN CMP_ROLE c ON a.role_id = c.role_id
WHERE c.group_id = v_group_id
GROUP BY c.role_id,c.subgroup,c.subgroup_description
ORDER BY c.subgroup;

同样，我不知道这是否比当前代码效率更高或更低。

如何优化这些oracle查询集

2 个答案: