Sql Oracle加速查询
此查询查找K2 isex的W_G:'L_E'被'E1'找到。 然后它找到K1 isex:“IS”是由“E1”找到的。 然后根据K1，K2'30'，10'，IS'和'C01A'找到K4，K5 要加速此查询，是否需要修改此查询 执行计划如下Sql Oracle加速查询,sql,oracle,query-performance,Sql,Oracle,Query Performance,此查询查找K2 isex的W_G:'L_E'被'E1'找到。 然后它找到K1 isex:“IS”是由“E1”找到的。 然后根据K1，K2'30'，10'，IS'和'C01A'找到K4，K5 要加速此查询，是否需要修改此查询 执行计划如下 ----------------------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | | 5 (0)| 00:00:01 |
| 1 | VIEW | | 1 | 13 | 2 (0)| 00:00:01 |
| 2 | TABLE ACCESS FULL | SYS_TEMP_0FD9D6617_CD798E | 1 | 26 | 2 (0)| 00:00:01 |
| 3 | TEMP TABLE TRANSFORMATION | | | | | |
| 4 | LOAD AS SELECT | SYS_TEMP_0FD9D6618_CD798E | | | | |
|* 5 | TABLE ACCESS FULL | C_W_B | 1 | 273 | 3 (0)| 00:00:01 |
| 6 | VIEW | | 1 | 13 | 2 (0)| 00:00:01 |
| 7 | TABLE ACCESS FULL | SYS_TEMP_0FD9D6617_CD798E | 1 | 26 | 2 (0)| 00:00:01 |
|* 8 | VIEW | | 1 | 39 | 2 (0)| 00:00:01 |
| 9 | TABLE ACCESS FULL | SYS_TEMP_0FD9D6618_CD798E | 1 | 273 | 2 (0)| 00:00:01 |
| 10 | SORT AGGREGATE | | 1 | 13 | | |
| 11 | VIEW | | 1 | 13 | 2 (0)| 00:00:01 |
| 12 | TABLE ACCESS FULL | SYS_TEMP_0FD9D6618_CD798E | 1 | 273 | 2 (0)| 00:00:01 |
| 13 | TEMP TABLE TRANSFORMATION | | | | | |
| 14 | LOAD AS SELECT | SYS_TEMP_0FD9D6617_CD798E | | | | |
|* 15 | TABLE ACCESS FULL | C_W_B | 1 | 78 | 3 (0)| 00:00:01 |
| 16 | FAST DUAL | | 1 | | 2 (0)| 00:00:01 |
----------------------------------------------------------------------------------------------------------
Statistics
----------------------------------------------------------
4 recursive calls
16 db block gets
30 consistent gets
2 physical reads
1156 redo size
682 bytes sent via SQL*Net to client
416 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
CREATE TABLE C_W_B (
HE VARCHAR2(22),
WE VARCHAR2(22),
G_D VARCHAR2(22),
W_G VARCHAR2(22),
W_V VARCHAR2(22),
K1 VARCHAR2(22),
K2 VARCHAR2(22),
K3 VARCHAR2(22),
K4 VARCHAR2(22),
K5 VARCHAR2(22),
K6 VARCHAR2(22),
K7 VARCHAR2(22),
K8 VARCHAR2(22),
K9 VARCHAR2(22),
K10 VARCHAR2(22),
K11 VARCHAR2(22),
K12 VARCHAR2(22),
K13 VARCHAR2(22),
K14 VARCHAR2(22),
K22 VARCHAR2(22),
K16 VARCHAR2(22)
);
INSERT INTO C_W_B(HE,WE,G_D,W_G,W_V,K1,K3,K4,K6) VALUES
('C6','61','R3','L_C_F','1','IS','C01A','30','10');
INSERT INTO C_W_B(HE,WE,G_D,W_G,K1,K2) VALUES
('C6','61','R3','L_E','IS','E1');
WITH G_S AS (
SELECT /*+FIRST_ROWS(1)*/
A.W_G, A.K1
FROM C_W_B A
WHERE 1=1
AND A.HE = 'C6'AND A.WE = '61' AND A.G_D ='R3' AND A.K2='E1'
)
SELECT CASE
WHEN (SELECT A.W_G FROM G_S A)='L_E'
THEN (
WITH SEO AS (
SELECT A.K1 FROM G_S A
),
SE AS (SELECT *
FROM C_W_B B
WHERE B.W_G = 'L_C_F'
AND B.HE = 'C6'
AND B.WE = '61'
AND B.G_D ='R3'
AND B.K1 = (SELECT K1 FROM SEO)
AND B.K3 LIKE 'C01A'
)
SELECT (L_T /QT*1000)
FROM ( SELECT NVL(TO_NUMBER(K4),0) AS L_T,
NVL(TO_NUMBER(K6),0) AS QT
FROM SE
WHERE W_V IN (SELECT TO_CHAR(MAX(TO_NUMBER(W_V))) FROM SE )
)
)
WHEN 1=2
THEN TO_NUMBER(0)
ELSE TO_NUMBER(0)
END
FROM DUAL;
您的查询可以简化，并可提供以下性能：
SELECT CASE
WHEN W_G ='L_E'
AND K6 IS NOT NULL -- ADDED K6 IS NOT NULL CONDITION TO AVOID NUMBER DIVIDE BY 0 ERROR
THEN NVL(K4,0) / (K6*1000)
ELSE 0
END
FROM (SELECT A.W_G, B.K6, B.K4,
ROW_NUMBER()
OVER (ORDER BY TO_NUMBER(W_V) DESC NULLS LAST) AS RN
FROM C_W_B A JOIN C_W_B S ON (A.K1=S.K1)
WHERE A.HE = 'C6'
AND A.WE = '61'
AND A.G_D ='R3'
AND A.K2='E1'
AND S.W_G = 'L_C_F'
AND S.HE = 'C6'
AND S.WE = '61'
AND S.G_D ='R3'
AND S.K3 LIKE 'C01A')
WHERE RN = 1;
干杯 现在这个查询的访问时间是多少？您的查询中有很多无用的东西。例如，对于\u CHARMAXTO\u NUMBERW\u V，为什么要用number和char环绕？显然CTE G_S只返回一行-用G_S替换dual。我个人不喜欢这些嵌入式子查询。也许试着把所有的表放在一个'FROM'子句中。我认为您不需要超过1-2个子查询。表的统计信息是最新的吗？