使用相同的表做比对后UPDATE,执行非常的慢

[LLJ_LILONGJIAN]

以下语句执行很慢，请问那位大牛帮忙看看，谢谢大家！
--前一个航班的降落时刻,航线,起飞城市,降落城市
UPDATE TMP_ET_TASK A
SET (PRIOR_DOWN_TIME
,PRIOR_FLIGHT_NO
,PRIOR_DOWN_CITY
,PRIOR_UP_CITY
,PRIOR_TASK_KIND
,PRIOR_TASK_ID) =
(SELECT B.LAG_DOWN_TIME
,B.LAG_FLIGHT_NO
,B.LAG_DOWN_CITY
,B.LAG_UP_CITY
,B.LAG_TASK_KIND
,B.LAG_TASK_ID
FROM V_TMP_ET_TASK_LAG B
WHERE B.TASK_ID = A.TASK_ID AND
B.MAN_ID = A.MAN_ID AND
ROWNUM = 1);

视图V_TMP_ET_TASK_LAG：
CREATE OR REPLACE VIEW V_TMP_ET_TASK_LAG AS
SELECT TASK_ID,MAN_ID,MAN_NAME, FACT_UP_TIME,FACT_DOWN_TIME, LINE_CODE, DOWN_CITY_name,
LAG(FLIGHT_NO) OVER(PARTITION BY MAN_ID ORDER BY FACT_UP_TIME) AS LAG_FLIGHT_NO,
LAG(FACT_DOWN_TIME) OVER(PARTITION BY MAN_ID ORDER BY FACT_UP_TIME) AS LAG_DOWN_TIME,
LAG(UP_CITY_name) OVER(PARTITION BY MAN_ID ORDER BY FACT_UP_TIME) AS LAG_UP_CITY,
LAG(DOWN_CITY_name) OVER(PARTITION BY MAN_ID ORDER BY FACT_UP_TIME) AS LAG_DOWN_CITY,
LAG(TASK_KIND) OVER(PARTITION BY MAN_ID ORDER BY FACT_UP_TIME) AS LAG_TASK_KIND,
LAG(TASK_ID) OVER(PARTITION BY MAN_ID ORDER BY FACT_UP_TIME) AS LAG_TASK_ID
FROM TMP_ET_TASK T;

[碧水幽幽泉]

试试这个：

UPDATE TMP_ET_TASK A
       SET (PRIOR_DOWN_TIME
          ,PRIOR_FLIGHT_NO
          ,PRIOR_DOWN_CITY
          ,PRIOR_UP_CITY
          ,PRIOR_TASK_KIND
          ,PRIOR_TASK_ID) =
           (SELECT B.LAG_DOWN_TIME
                  ,B.LAG_FLIGHT_NO
                  ,B.LAG_DOWN_CITY
                  ,B.LAG_UP_CITY
                  ,B.LAG_TASK_KIND
                  ,B.LAG_TASK_ID
              FROM V_TMP_ET_TASK_LAG B
             WHERE B.TASK_ID = A.TASK_ID AND
                   B.MAN_ID = A.MAN_ID AND
                   ROWNUM = 1)
WHERE EXISTS(  SELECT NULL 
              FROM V_TMP_ET_TASK_LAG B
             WHERE B.TASK_ID = A.TASK_ID AND
                   B.MAN_ID = A.MAN_ID AND
                   );

[sxq129601]

因为你没加WHERE条件呀，相当于全表都更新了一遍