On Thu, 3 Apr 2008, Tom Lane wrote:
> As of PG 8.3, the bgwriter tries very hard to make the elapsed time of a
> checkpoint be just about checkpoint_timeout *
> checkpoint_completion_target, regardless of load factors.
In the cases where the timing on checkpoint writes are timeout driven.
When they're checkpoint_segments driven the time is derived from that
drumbeat instead. And those cases are the more interesting ones, because
those are the ones that will speed up during really busy periods.
I think it's completely relevant to track how often checkpoints are
happening because they do still vary even in 8.3. I'm just not sure if
the current methods available for that really aren't good enough, or if
it's just the case that not everyone is aware of all of them.
--
* Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD