2. 程式人生 > 其它 >2表關聯更新測試

2表關聯更新測試

阿新 • • 發佈：2022-05-26

第一種

UPDATE yz.T1 a
SET a.object_name =
(select b.object_name from yz.t2 b where a.object_id = b.object_id)
WHERE EXISTS (SELECT 1 FROM yz.t2 b WHERE b.object_id = a.object_id);

缺點：

  1、執行時間長，耗費資源

  2、表中object_id列必須唯一

SQL> UPDATE yz.T1 a
   SET a.object_name =
       (select b.object_name from yz.t2 b where a.object_id = b.object_id)
 WHERE EXISTS (SELECT 1 FROM yz.t2 b WHERE b.object_id = a.object_id);  2    3    4  

4977 rows updated.

Elapsed: 00:07:22.01

[qdtais1]@ht19c01[/home/oracle]$./ora sql 6fd93n7jb1jkg

DBA_HIST_SQLSTAT detail(recent 15 days):


 SNAP_ID SNAP_DATE	      INST	  PHV	CPU_S	  EXECS     BUF_PE     ROW_PE	  MS_PE     DISK_R    IO_S     APP_MS
-------- -------------------- ---- ---------- ------- --------- ---------- ---------- --------- ---------- ------- ----------
    1095 20220525 18:01 	 1 3641560594	  420	      1  432536652	 4977	 439046  432466359     205	    0


SQL top event in gv_active_session_history:

   INST_ID EVENT						     CNT
---------- -------------------------------------------------- ----------
	 1							      94
	 1 direct path read					      91
	 1 db file scattered read				       5



Historical SQL plans in AWR:


PLAN_TABLE_OUTPUT
------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID 6fd93n7jb1jkg
--------------------
UPDATE yz.T1 a	  SET a.object_name =	     (select b.object_name from
yz.t2 b where a.object_id = b.object_id)  WHERE EXISTS (SELECT 1 FROM
yz.t2 b WHERE b.object_id = a.object_id)

Plan hash value: 3641560594

------------------------------------------------------------------------------
| Id  | Operation	      | Name | Rows  | Bytes | Cost (%CPU)| Time     |
------------------------------------------------------------------------------
|   0 | UPDATE STATEMENT      |      |	     |	     |	 177M(100)|	     |
|   1 |  UPDATE 	      | T1   |	     |	     |		  |	     |
|   2 |   HASH JOIN RIGHT SEMI|      |	4977 |	 184K| 54130   (1)| 00:00:03 |
|   3 |    TABLE ACCESS FULL  | T2   |	4977 | 19908 | 23748   (1)| 00:00:01 |
|   4 |    TABLE ACCESS FULL  | T1   |	9975 |	 331K| 30382   (1)| 00:00:02 |
|   5 |   TABLE ACCESS FULL   | T2   |	   1 |	  23 | 23748   (1)| 00:00:01 |
------------------------------------------------------------------------------




Current SQL plans in Curor:


PLAN_TABLE_OUTPUT
------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID	6fd93n7jb1jkg, child number 0
-------------------------------------
UPDATE yz.T1 a	  SET a.object_name =	     (select b.object_name from
yz.t2 b where a.object_id = b.object_id)  WHERE EXISTS (SELECT 1 FROM
yz.t2 b WHERE b.object_id = a.object_id)

Plan hash value: 3641560594

-------------------------------------------------------------------------------
| Id  | Operation	      | Name | E-Rows |E-Bytes| Cost (%CPU)| E-Time   |
-------------------------------------------------------------------------------
|   0 | UPDATE STATEMENT      |      |	      |       |   177M(100)|	      |
|   1 |  UPDATE 	      | T1   |	      |       | 	   |	      |
|*  2 |   HASH JOIN RIGHT SEMI|      |	 4977 |   184K| 54130	(1)| 00:00:03 |
|   3 |    TABLE ACCESS FULL  | T2   |	 4977 | 19908 | 23748	(1)| 00:00:01 |
|   4 |    TABLE ACCESS FULL  | T1   |	 9975 |   331K| 30382	(1)| 00:00:02 |
|*  5 |   TABLE ACCESS FULL   | T2   |	    1 |    23 | 23748	(1)| 00:00:01 |
-------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   2 - access("B"."OBJECT_ID"="A"."OBJECT_ID")
   5 - filter("B"."OBJECT_ID"=:B1)

Note
-----
   - Warning: basic plan statistics not available. These are only collected when:
       * hint 'gather_plan_statistics' is used for the statement or
       * parameter 'statistics_level' is set to 'ALL', at session or system level



Historical Plans Summary(dba_hist_sqlstat):

  RN PLAN_HASH_VALUE AVG_ETIME_S  AVG_CPU_S    AVG_BUFFERS  AVG_READS	AVG_ROWS     EXECS FIRST_SNAP	       LAST_SNAP
---- --------------- ------------ ------------ ----------- ---------- ---------- --------- ------------------- -------------------
   1	  3641560594	  439.046      420.428	 432536652  432466359	    4977	 1 2022-05-25 18:01:46 2022-05-25 18:01:46


Current Plans Summary(gv_sql):

 RN PLAN_HASH_VALUE AVG_ETIME_S  AVG_CPU_S    AVG_BUFFERS  AVG_READS   AVG_ROWS TOTAL_EXEC FIRST_LOAD_TIME	LAST_ACTIVE
--- --------------- ------------ ------------ ----------- ---------- ---------- ---------- -------------------- --------------------
  1	 3641560594	 193.348      178.332	183979908  183950468	      0 	 1 2022-05-26/11:02:59	2022-05-26 11:06:12


Tables involved(used objects may not exists now):


TABLE_OWNER		       TABLE_NAME			NUM_ROWS LAST_ANALYZED		  SIZE_M
------------------------------ ------------------------------ ---------- -------------------- ----------
YZ			       T2				    4977 2022-05-25 17:47:09	     688
YZ			       T1				    9975 2022-05-25 17:47:08	     880


Above is sql info for sql_id :6fd93n7jb1jkg ,plan format is typical. for advanced plan ,use ora sql <sqlid> adv

　　清空shared pool 和 db cache  

alter system flush buffer_cache;
alter system flush shared_pool;

　　使用merge更新，merge 更新時表中的資料可以不時唯一的

merge into yz.t1 a
using (select t2.object_id,t2.object_name from yz.t2) b
on (a.object_id = b.object_id)
when matched then
update set a.object_name = b.object_name;

SQL> merge into yz.t1 a
using (select t2.object_id,t2.object_name from yz.t2) b
on (a.object_id = b.object_id)
when matched then 
  update  set a.object_name = b.object_name;  2    3    4    5  

4977 rows merged.

Elapsed: 00:00:00.52
[qdtais1]@ht19c01[/home/oracle]$./ora sql 337s73614yhbj

DBA_HIST_SQLSTAT detail(recent 15 days):


 SNAP_ID SNAP_DATE	      INST	  PHV	CPU_S	  EXECS     BUF_PE     ROW_PE	  MS_PE     DISK_R    IO_S     APP_MS
-------- -------------------- ---- ---------- ------- --------- ---------- ---------- --------- ---------- ------- ----------
    1096 20220526 11:12 	 1 2683531971	    0	      1     204686	 4977	    415     198825	 0	    0


SQL top event in gv_active_session_history:



Historical SQL plans in AWR:


PLAN_TABLE_OUTPUT
------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID 337s73614yhbj
--------------------
merge into yz.t1 a using (select t2.object_id,t2.object_name from
yz.t2) b on (a.object_id = b.object_id) when matched then   update  set
a.object_name = b.object_name

Plan hash value: 2683531971

-----------------------------------------------------------------------------
| Id  | Operation	     | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | MERGE STATEMENT      |	    |	    |	    | 54130 (100)|	    |
|   1 |  MERGE		     | T1   |	    |	    |		 |	    |
|   2 |   VIEW		     |	    |	    |	    |		 |	    |
|   3 |    HASH JOIN	     |	    |  4977 |	816K| 54130   (1)| 00:00:03 |
|   4 |     TABLE ACCESS FULL| T2   |  4977 |	111K| 23748   (1)| 00:00:01 |
|   5 |     TABLE ACCESS FULL| T1   |  9975 |  1412K| 30383   (1)| 00:00:02 |
-----------------------------------------------------------------------------




Current SQL plans in Curor:


PLAN_TABLE_OUTPUT
------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID	337s73614yhbj, child number 0
-------------------------------------
merge into yz.t1 a using (select t2.object_id,t2.object_name from
yz.t2) b on (a.object_id = b.object_id) when matched then   update  set
a.object_name = b.object_name

Plan hash value: 2683531971

---------------------------------------------------------------------------------------------------------
| Id  | Operation	     | Name | E-Rows |E-Bytes| Cost (%CPU)| E-Time   |	OMem |	1Mem | Used-Mem |
---------------------------------------------------------------------------------------------------------
|   0 | MERGE STATEMENT      |	    |	     |	     | 54130 (100)|	     |	     |	     |		|
|   1 |  MERGE		     | T1   |	     |	     |		  |	     |	     |	     |		|
|   2 |   VIEW		     |	    |	     |	     |		  |	     |	     |	     |		|
|*  3 |    HASH JOIN	     |	    |	4977 |	 816K| 54130   (1)| 00:00:03 |	1355K|	1355K| 1662K (0)|
|   4 |     TABLE ACCESS FULL| T2   |	4977 |	 111K| 23748   (1)| 00:00:01 |	     |	     |		|
|   5 |     TABLE ACCESS FULL| T1   |	9975 |	1412K| 30383   (1)| 00:00:02 |	     |	     |		|
---------------------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   3 - access("A"."OBJECT_ID"="T2"."OBJECT_ID")

Note
-----
   - Warning: basic plan statistics not available. These are only collected when:
       * hint 'gather_plan_statistics' is used for the statement or
       * parameter 'statistics_level' is set to 'ALL', at session or system level



Historical Plans Summary(dba_hist_sqlstat):

  RN PLAN_HASH_VALUE AVG_ETIME_S  AVG_CPU_S    AVG_BUFFERS  AVG_READS	AVG_ROWS     EXECS FIRST_SNAP	       LAST_SNAP
---- --------------- ------------ ------------ ----------- ---------- ---------- --------- ------------------- -------------------
   1	  2683531971	    0.415	 0.410	    204686     198825	    4977	 1 2022-05-26 11:12:11 2022-05-26 11:12:11


Current Plans Summary(gv_sql):

 RN PLAN_HASH_VALUE AVG_ETIME_S  AVG_CPU_S    AVG_BUFFERS  AVG_READS   AVG_ROWS TOTAL_EXEC FIRST_LOAD_TIME	LAST_ACTIVE
--- --------------- ------------ ------------ ----------- ---------- ---------- ---------- -------------------- --------------------
  1	 2683531971	   0.521	0.470	   205276     199000	   4977 	 1 2022-05-26/11:15:26	2022-05-26 11:15:27


Tables involved(used objects may not exists now):


TABLE_OWNER		       TABLE_NAME			NUM_ROWS LAST_ANALYZED		  SIZE_M
------------------------------ ------------------------------ ---------- -------------------- ----------
YZ			       T2				    4977 2022-05-25 17:47:09	     688
YZ			       T1				    9975 2022-05-25 17:47:08	     880


Above is sql info for sql_id :337s73614yhbj ,plan format is typical. for advanced plan ,use ora sql <sqlid> adv

　　

 通過上述資料分析，可知使用merge更新比傳統的update執行事件快1000倍

相關推薦

2表關聯更新測試

第一種 UPDATE yz.T1 aSET a.object_name =(select b.object_name from yz.t2 b where a.object_id = b.object_id) WHERE EXISTS (SELECT 1 FROM yz.t2 b WHERE b.object_id = a.object_id);

SQL update 多表關聯更新的實現程式碼

實現多表更新，尤其是A表和A的子表B表資料更新，下面是例子 有A、B張表，其記錄如下：

【 DB_Oracle】Oracle多表關聯更新

很多場景我們需要根據多個表的某欄位進行關聯更新。   select * from table1  t1;

postgresql 兩表關聯更新操作

我就廢話不多說了，大家還是直接看程式碼吧~ UPDATE 要更新的表 SET 欄位1 = cqt.欄位1,欄位2 = cqt.欄位2,FROM

Oracle中多表關聯批量插入批量更新與批量刪除操作

該文章會分為三部分 1.多表關聯批量插入 2.多表關聯批量更新 3.多表關聯批量刪除

英雄聯盟手遊 2.2 版本即將更新：極地大亂鬥開啟測試

3月23日訊息《英雄聯盟手遊》 2.2 版本即將更新，官方在更新日誌中表示卡茲克、雷恩加爾等英雄將在新版本上線，休閒玩家喜歡的極地大亂鬥模式也將開啟測試。

《腐爛國度2》提升敵人難度更新測試中

《腐爛國度2》正在改變殭屍群落的行為方式，讓這些不死生物能夠對玩家的社群造成真正威脅。

資料庫的建立方法 - 2張表 -帶表關聯

　　1.　　首先建立資料 , 建立表  　　2.　　檔案model.py程式碼:   from django.db import models\'\'\'部門表\'\'\'class BuMen(models.Model):title = models.CharField(verbose_name=\'部門名稱\', max_leng

MySQL之多表關聯刪除/更新

日常測試的時候，需要連線其他表而刪除某些髒資料，按照正常的查詢的寫法，會這樣寫刪除語句：

Swoft 2.0.5 更新，新增高效秒級定時任務、異常管理元件

什麼是 Swoft ? Swoft 是一款基於 Swoole 擴充套件實現的 PHP 微服務協程框架。Swoft 能像 Go 一樣，內建協程網路伺服器及常用的協程客戶端且常駐記憶體，不依賴傳統的 PHP-FPM。有類似 Go 語言的協程操作方式，有類

mysql實現多表關聯統計(子查詢統計)示例

本文例項講述了mysql實現多表關聯統計的方法。分享給大家供大家參考，具體如下：

MongoDB多表關聯查詢操作例項詳解

本文例項講述了MongoDB多表關聯查詢操作。分享給大家供大家參考，具體如下：

MongoDB中多表關聯查詢（$lookup）的深入講解

一. 聚合框架 聚合框架是MongoDB的高階查詢語言，它允許我們通過轉換和合並多個文件中的資料來生成新的單個文件中不存在的資訊。

Mysql 根據一個表資料更新另一個表的某些欄位(sql語句)

下面一段程式碼給大家介紹Mysql 根據一個表資料更新另一個表的某些欄位，具體程式碼如下所示：

Mysql update多表聯合更新的方法小結

下面我建兩個表，並執行一系列sql語句，仔細觀察sql執行後表中資料的變化，很容易就能理解多表聯合更新的用法

Spring boot2基於Mybatis實現多表關聯查詢

模擬業務關係： 一個使用者user有對應的一個公司company，每個使用者有多個賬戶account。

MySQL 多表關聯一對多查詢實現取最新一條資料的方法示例

本文例項講述了MySQL 多表關聯一對多查詢實現取最新一條資料的方法。分享給大家供大家參考，具體如下：

Spring Boot入門系列（十七）整合Mybatis，建立自定義mapper 實現多表關聯查詢！

之前講了Springboot整合Mybatis，介紹瞭如何自動生成pojo實體類、mapper類和對應的mapper.xml 檔案，並實現最基本的增刪改查功能。mybatis 外掛自動生成的mapper 實現了大部分基本、通用的方法，如：insert、update、

Mybatis xxxMapper.xml 三表關聯，配置檔案

VideoMapper.xml <?xml version=\"1.0\" encoding=\"UTF-8\" ?> <!DOCTYPE mapper PUBLIC \"-//mybatis.org//DTD Mapper 3.0//EN\"

詳解MongoDB中的多表關聯查詢（$lookup）

詳解MongoDB中的多表關聯查詢（$lookup） 作者：東山絮柳仔 本文轉載於：https://www.cnblogs.com/xuliuzai/p/10055535.html