From: Jan Kirchhoff (kirchygmx.de)
Date: Tue Jan 29 2008 - 17:09:51 CST

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

David Schneider-Joseph schrieb:
> Hi all,
>
> I am attempting to convert a very large table (~23 million rows) from
> MyISAM to InnoDB. If I do it in chunks of one million at a time, the
> first million are very fast (approx. 3 minutes or so), and then it
> gets progressively worse, until by the time I get even to the fourth
> chunk, it's taking 15-20 minutes, and continuing to worsen. This is
> much worse degradation than the O*log(N) that you would expect.
>
[...]
>
> This problem can even be reproduced in a very simple test case, where
> I continuously insert approximately 1 million rows into a table, with
> random data. `big_table` can be any table with approximately one
> million rows in id range 1 through 1000000 (we're not actually using
> any data from it):
>
[...]
> Any ideas, anyone?
>
> Thanks,
> David
>

what hardware are you running on and you much memory do you have? what
version of mysql?| |
How did you set innodb_buffer_pool_size? you might want to read
http://dev.mysql.com/doc/refman/5.0/en/innodb-tuning.html
and do some tuning.

In case that doesn't help you, you'll need to post more info on your config.

Jan

--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]