Comments on: What do the InnoDB insert buffer statistics mean? http://www.xaprb.com/blog/2009/10/25/what-do-the-innodb-insert-buffer-statistics-mean/ Stay curious! Thu, 02 May 2013 12:36:53 +0000 hourly 1 http://wordpress.org/?v=3.5.1 By: Log Buffer #167: a Carnival of the Vanities for DBAs | Pythian Group Blog http://www.xaprb.com/blog/2009/10/25/what-do-the-innodb-insert-buffer-statistics-mean/#comment-17190 Log Buffer #167: a Carnival of the Vanities for DBAs | Pythian Group Blog Fri, 30 Oct 2009 17:04:41 +0000 http://www.xaprb.com/blog/?p=1392#comment-17190 [...] Schwartz also has a question for you, what do the InnoDB insert buffer statistics mean?: Ever seen this in SHOW INNODB STATUS and wondered what it means? [...]

]]>
By: Mark Callaghan http://www.xaprb.com/blog/2009/10/25/what-do-the-innodb-insert-buffer-statistics-mean/#comment-17171 Mark Callaghan Wed, 28 Oct 2009 01:45:59 +0000 http://www.xaprb.com/blog/?p=1392#comment-17171 The code I copied is from ibuf_data_sizes_update and cscope shows it is called from ibuf_data_init_for_space, ibuf_insert_low and ibuf_delete_rec. I assume it is called after init.

]]>
By: Xaprb http://www.xaprb.com/blog/2009/10/25/what-do-the-innodb-insert-buffer-statistics-mean/#comment-17170 Xaprb Wed, 28 Oct 2009 01:42:15 +0000 http://www.xaprb.com/blog/?p=1392#comment-17170 s/believe it is/believe it does/g

]]>
By: Xaprb http://www.xaprb.com/blog/2009/10/25/what-do-the-innodb-insert-buffer-statistics-mean/#comment-17169 Xaprb Wed, 28 Oct 2009 01:41:56 +0000 http://www.xaprb.com/blog/?p=1392#comment-17169 Right, I saw the same code — but I think that code is only executed when the insert buffer is initialized, and I wanted to additionally understand whether it holds as the insert buffer is modified over time, which I believe it is.

]]>
By: Mark Callaghan http://www.xaprb.com/blog/2009/10/25/what-do-the-innodb-insert-buffer-statistics-mean/#comment-17168 Mark Callaghan Wed, 28 Oct 2009 01:34:58 +0000 http://www.xaprb.com/blog/?p=1392#comment-17168 I copied that from the code, so it should hold. The other interesting number is:
(‘merged records’ – ‘merges’) * 2
That is an estimate for the IO (reads+writes) saved by the insert buffer.

]]>