How to remove duplicates in MySql table with 29 million rows?

Question

MySQL

What request will the system not get up for me during such an operation?

Answer 1

If it is possible to stop writing/changing the table, the fastest method would be the following scheme.
1.
Create a new table table2 with a structure similar to the original table1.
2.

insert into table2 (select distinct ... from table 1);

if the structure is complex, you can use group by
3 instead of distinct.
4.
Rename table2 to table1;

Answer 2

If I were you, I wouldn't delete anything from the table. And I would simply create a new one, already without duplicates. In terms of resource consumption, it will be much more economical. And no one bothers to do this:
1. Selected the first hundred thousand, remember the id of the last entry.
2. Pause.
3. Selected the next hundred thousand starting from the last ID. Go that point two.
All.

Answer 3

Have a look - stackoverflow.com/questions/4685173/delete-all-dup...
stackoverflow.com/questions/3311903/remove-duplica...
stackoverflow.com/questions/14046355/how-do-i-dele...