批量UPDATE的操作

有乙個簡單的業務需求，要根據另外乙個表的id去更新這張表的sys為scott；要更新２１１w記錄。

sql> create index t_idx1 on t_merge1(object_id);
index created.
sql> create index idx_t on t_merge2(object_id);
index created.
update t_merge1 set owner='scott' where object_id in(select object_id from t_merge2 where t_merge2.object_id=t_merge2.object_id);
21047392 rows updated.
elapsed: 00:32:16.39
耗時：　３２分鐘
merge into t_merge1 a 
using (select object_id from t_merge2) b
on(a.object_id=b.object_id)
when matched then 
update set a.owner='scott';
21047392 rows merged.
elapsed: 00:13:22.77
耗時：１３分鐘

merge的操作大大提高了更新速度。

批量插入，update

setting 1 create table t as select from all objects where 1 2 模擬逐行提交的情況，注意觀察執行時間 declare begin for cur in select from t ref loop insert into t values ...

mysql批量更新update操作，導致鎖表

sql如下 update pay trans set return state 1 where order id and user id 批量執行此條sql時，導致資料表被鎖，所有交易異常。經排查where條件order id and user id 沒有加索引，導致整個表被鎖。mysql的行鎖是通...

Oracle批量Update記錄

工作中經常用到oracle批量更新記錄，做為老手也怕出錯，總之要小心再小心，確保資料批量更新正確。下面舉乙個例子 1 建立兩張結構類似的表，建表語句如下 create table jayt1 id int,code varchar2 8 create table jayt2 id int,code ...