160802, 1.06s delete 10w duplicate record only keep one (MySQL) and Linux deleted garbled file

Last published once deleted duplicate records of the article, where the best solution three deleted 1w also took 0.07 seconds and 2w has been spent 4 seconds. Further optimizations were made today, and it took only 1.06 seconds to test the removal of the 10w bar. The speed has improved a lot.

Build a Table statement

CREATE TABLE ' Test_user ' (

' ID ' INT (Ten) not NULL DEFAULT ' 0 ',

' Namea ' VARCHAR (+) not NULL,

' Nameb ' VARCHAR (+) not NULL,

' user_id ' VARCHAR (+) not NULL,

PRIMARY KEY (' id ')

) Engine=innodb DEFAULT Charset=utf8;

New scenario: (where user_id is a repeating field, keep one)

DELETE from Test_user WHERE ID not in (select-ID from (select MAX (ID) as-ID from Test_user GROUP by user_id) X);

10,000 Delete 9990 20,000 delete 19990 100,000 strip delete 199990
0.062sec 0.018sec 1.060sec

In addition, we summarize a method to delete Linux garbled files: (where the red part is the Inode (index node), the-I option is used to display the inode)
[[email protected] test]# ls
?????. Txt[[email protected] test]# Ls-lih
Total 4.0K
1061229-rw-r--r--. 1 root root 104 Jul 20:16?????. Txt

[[email protected] test]# find. -inum 1061229-exec rm {}-RF \;

[Email protected] test]# ls-l
Total 0

160802, 1.06s delete 10w duplicate record only keep one (MySQL) and Linux deleted garbled file