替换SQL执行计划

Switching two different SQL Plan with SQL Profile in Oracle...

当SQL是业务系统动态生成的，或者是第三方系统产生的，在数据库层面分析发现性能问题时，可能难以实现及时修改业务程序改善执行计划和性能；但可以在数据库层面找到有问题的SQL，调整和改造该SQL，然后将执行计划应用到原始的SQL语句中，步骤如下:

• 在数据库层面找到性能问题的SQL相关信息;
• 重构和优化SQL;
• 对比旧的和优化后的SQL性能信息;
• 将最优的SQL执行计划应用到原始SQL语句上;

通过这样的思路，在数据库层面优化SQL性能。来看看下面的示例；

SQL> @i

USERNAME             INST_NAME            HOST_NAME                 SID   SERIAL#  VERSION    STARTED  SPID       OPID  CPID            SADDR            PADDR
-------------------- -------------------- ------------------------- ----- -------- ---------- -------- ---------- ----- --------------- ---------------- ----------------
OPS$SYWU             sydb                 sywu.com                  154   23601    11.2.0.4.0 20160421 23407      25    4144:3660       0000000071E1CC20 0000000072134028

有下面的2个SQL，SQL 1;

select
    e.first_name,e.email,e.salary,d.department_name,j.job_title
from
    employees e,departments d,jobs j
where
    e.department_id=d.department_id and e.job_id=j.job_id
    and e.salary>8500
order by
    j.job_title
;

-------------------------------------------------------------------------------------------------------------------------------------------------------------------
| Id  | Operation                      | Name        | Starts | E-Rows |E-Bytes| Cost (%CPU)| E-Time   | A-Rows |   A-Time   | Buffers |  OMem |  1Mem | Used-Mem |
-------------------------------------------------------------------------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT               |             |      1 |        |       |    40 (100)|          |     29 |00:00:00.01 |      44 |       |       |          |
|   1 |  SORT ORDER BY                 |             |      1 |     28 |  2072 |    40   (5)| 00:00:01 |     29 |00:00:00.01 |      44 |  4096 |  4096 | 4096  (0)|
|*  2 |   HASH JOIN                    |             |      1 |     28 |  2072 |    39   (3)| 00:00:01 |     29 |00:00:00.01 |      44 |   600K|   600K|  802K (0)|
|   3 |    MERGE JOIN                  |             |      1 |     28 |  1624 |    21   (5)| 00:00:01 |     29 |00:00:00.01 |      23 |       |       |          |
|   4 |     TABLE ACCESS BY INDEX ROWID| JOBS        |      1 |     19 |   513 |     2   (0)| 00:00:01 |     17 |00:00:00.01 |       2 |       |       |          |
|   5 |      INDEX FULL SCAN           | JOB_ID_PK   |      1 |     19 |       |     1   (0)| 00:00:01 |     17 |00:00:00.01 |       1 |       |       |          |
|*  6 |     SORT JOIN                  |             |     17 |     28 |   868 |    19   (6)| 00:00:01 |     29 |00:00:00.01 |      21 |  2048 |  2048 | 2048  (0)|
|*  7 |      TABLE ACCESS FULL         | EMPLOYEES   |      1 |     28 |   868 |    18   (0)| 00:00:01 |     29 |00:00:00.01 |      21 |       |       |          |
|   8 |    TABLE ACCESS FULL           | DEPARTMENTS |      1 |     27 |   432 |    18   (0)| 00:00:01 |     27 |00:00:00.01 |      21 |       |       |          |
-------------------------------------------------------------------------------------------------------------------------------------------------------------------

SQL 1是原始SQL；下面是调整后的SQL 2;

select /*+ ordered index(d DEPT_ID_PK) index(j JOB_ID_PK) */
   e.first_name,e.email,e.salary,d.department_name,j.job_title
from
   employees e,departments d,jobs j
where
   e.department_id=d.department_id and e.job_id=j.job_id
   and e.salary>8500
order by
   j.job_title
;

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
| Id  | Operation                       | Name              | Starts | E-Rows |E-Bytes| Cost (%CPU)| E-Time   | A-Rows |   A-Time   | Buffers | Reads  |  OMem |  1Mem | Used-Mem |
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT                |                   |      1 |        |       |     8 (100)|          |     29 |00:00:00.02 |       6 |      2 |       |       |          |
|   1 |  SORT ORDER BY                  |                   |      1 |     28 |  2072 |     8  (25)| 00:00:01 |     29 |00:00:00.02 |       6 |      2 |  4096 |  4096 | 4096  (0)|
|*  2 |   HASH JOIN                     |                   |      1 |     28 |  2072 |     7  (15)| 00:00:01 |     29 |00:00:00.02 |       6 |      2 |   639K|   639K|  881K (0)|
|   3 |    MERGE JOIN                   |                   |      1 |     28 |  1316 |     5  (20)| 00:00:01 |     29 |00:00:00.02 |       4 |      2 |       |       |          |
|*  4 |     TABLE ACCESS BY INDEX ROWID | EMPLOYEES         |      1 |     28 |   868 |     2   (0)| 00:00:01 |     29 |00:00:00.01 |       2 |      1 |       |       |          |
|   5 |      INDEX FULL SCAN            | EMP_DEPARTMENT_IX |      1 |    106 |       |     1   (0)| 00:00:01 |    106 |00:00:00.01 |       1 |      1 |       |       |          |
|*  6 |     SORT JOIN                   |                   |     29 |     27 |   432 |     3  (34)| 00:00:01 |     29 |00:00:00.01 |       2 |      1 |  2048 |  2048 | 2048  (0)|
|   7 |      TABLE ACCESS BY INDEX ROWID| DEPARTMENTS       |      1 |     27 |   432 |     2   (0)| 00:00:01 |     27 |00:00:00.01 |       2 |      1 |       |       |          |
|   8 |       INDEX FULL SCAN           | DEPT_ID_PK        |      1 |     27 |       |     1   (0)| 00:00:01 |     27 |00:00:00.01 |       1 |      1 |       |       |          |
|   9 |    TABLE ACCESS BY INDEX ROWID  | JOBS              |      1 |     19 |   513 |     2   (0)| 00:00:01 |     19 |00:00:00.01 |       2 |      0 |       |       |          |
|  10 |     INDEX FULL SCAN             | JOB_ID_PK         |      1 |     19 |       |     1   (0)| 00:00:01 |     19 |00:00:00.01 |       1 |      0 |       |       |          |
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

需要做的是在不调整和修改系统代码的情况下使原始的SQL 1使用第二个调整后(SQL 2)的执行计划，so 我们通过创建SQL profile来实现这个目的;

def v_sqlid='01h5fh3dhccyf'

declare
    l_sql clob;
begin
    select t.SQL_FULLTEXT into l_sql from v$sql t where sql_id='&v_sqlid';
    dbms_sqltune.import_sql_profile(sql_text => l_sql,name => 'pro_&v_sqlid',profile => sqlprof_attr('
IGNORE_OPTIM_EMBEDDED_HINTS
OPTIMIZER_FEATURES_ENABLE(''11.2.0.4'')
DB_VERSION(''11.2.0.4'')
ALL_ROWS
OUTLINE_LEAF(@"SEL$1")
INDEX(@"SEL$1" "E"@"SEL$1" ("EMPLOYEES"."DEPARTMENT_ID"))
INDEX(@"SEL$1" "D"@"SEL$1" ("DEPARTMENTS"."DEPARTMENT_ID"))
INDEX(@"SEL$1" "J"@"SEL$1" ("JOBS"."JOB_ID"))
LEADING(@"SEL$1" "E"@"SEL$1" "D"@"SEL$1" "J"@"SEL$1")
USE_MERGE(@"SEL$1" "D"@"SEL$1")
USE_HASH(@"SEL$1" "J"@"SEL$1")'
    ));
    dbms_output.put_line('SQL Profile:pro_&v_sqlid imported...');
end;
/

SQL Profile:pro_01h5fh3dhccyf imported...

PL/SQL procedure successfully completed.

当重新加载SQL时，执行计划改变，SQL profile被使用;

select
    e.first_name,e.email,e.salary,d.department_name,j.job_title
from
    employees e,departments d,jobs j
where
    e.department_id=d.department_id and e.job_id=j.job_id
    and e.salary>8500
order by
    j.job_title
;

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
| Id  | Operation                       | Name              | Starts | E-Rows |E-Bytes| Cost (%CPU)| E-Time   | A-Rows |   A-Time   | Buffers | Reads  |  OMem |  1Mem | Used-Mem |
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT                |                   |      1 |        |       |     8 (100)|          |     29 |00:00:00.02 |       6 |      2 |       |       |          |
|   1 |  SORT ORDER BY                  |                   |      1 |     28 |  2072 |     8  (25)| 00:00:01 |     29 |00:00:00.02 |       6 |      2 |  4096 |  4096 | 4096  (0)|
|*  2 |   HASH JOIN                     |                   |      1 |     28 |  2072 |     7  (15)| 00:00:01 |     29 |00:00:00.02 |       6 |      2 |   639K|   639K|  869K (0)|
|   3 |    MERGE JOIN                   |                   |      1 |     28 |  1316 |     5  (20)| 00:00:01 |     29 |00:00:00.01 |       4 |      2 |       |       |          |
|*  4 |     TABLE ACCESS BY INDEX ROWID | EMPLOYEES         |      1 |     28 |   868 |     2   (0)| 00:00:01 |     29 |00:00:00.01 |       2 |      1 |       |       |          |
|   5 |      INDEX FULL SCAN            | EMP_DEPARTMENT_IX |      1 |    106 |       |     1   (0)| 00:00:01 |    106 |00:00:00.01 |       1 |      1 |       |       |          |
|*  6 |     SORT JOIN                   |                   |     29 |     27 |   432 |     3  (34)| 00:00:01 |     29 |00:00:00.01 |       2 |      1 |  2048 |  2048 | 2048  (0)|
|   7 |      TABLE ACCESS BY INDEX ROWID| DEPARTMENTS       |      1 |     27 |   432 |     2   (0)| 00:00:01 |     27 |00:00:00.01 |       2 |      1 |       |       |          |
|   8 |       INDEX FULL SCAN           | DEPT_ID_PK        |      1 |     27 |       |     1   (0)| 00:00:01 |     27 |00:00:00.01 |       1 |      1 |       |       |          |
|   9 |    TABLE ACCESS BY INDEX ROWID  | JOBS              |      1 |     19 |   513 |     2   (0)| 00:00:01 |     19 |00:00:00.01 |       2 |      0 |       |       |          |
|  10 |     INDEX FULL SCAN             | JOB_ID_PK         |      1 |     19 |       |     1   (0)| 00:00:01 |     19 |00:00:00.01 |       1 |      0 |       |       |          |
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Note
-----
   - SQL profile pro_01h5fh3dhccyf used for this statement

原始的SQL使用了调整后的执行计划。