删除上亿级别时,注意回滚表空间

这次在做CSOM的masking的时候(级别低啊,只能做些简单的体力活,郁闷,要努力!)

遇到一个table,有几亿条record,import的过程中,报错:buffer超过,后来是statistics(pls refer the pervious note )

---------------------------------------

table的version date is from 20070331---20140331(monthly job),  I  just needed the lastest version date data,

should delete from the data ver_dte<to_date('20140301','yyyymmdd'),

But,  at this time, 回滚表空间的问题出来了!!!

几亿条record,占据大量的空间, delete from是完全行不通。。。 一个月一个月的删,超级慢。。

so I asked for 度娘,佢said I can create a new temp table which store the data I needed from the source table .

then truncate the source table ....

insert into source_table

select * from temp_table

Done !

猜你喜欢

转载自ticojj.iteye.com/blog/2071622