Re: New EXPLAIN ANALYZE statement

On Sat, Jul 21, 2001 at 02:18:45AM +1000, Justin Clift wrote:
> Hi Martijn,
>
> Why would you want to suppress the actual effect of non-select
> statements?
>
> If people want to time them, I think they should.  If they don't, they
> shouldn't run the query.  :)

Well, it depends if most of the time is taken working out what to enter or
if it's taken actually inserting the data (for an INSERT statement). This
patch is more aimed toward helping optimise the query, and just like the
planner doesn't take into account the time taken to send the data to the
frontend, this doesn't care how long it takes to insert the data, since that
is not something you can change.

If you want overall resource usage and the time taken to execute the whole
query, that already exists. If you want to find bottlenecks in the
processing then you need to get a profiler. This patch is aimed at the
planner because that's the *only* area that can be tuned and adjusted by
people trying to improve performance of thier queries.

> Is it harder to create or manage?

To change the way this code is called is easy, just fiddle the grammer. Once
people agree on a name it'll be done.

--
Martijn van Oosterhout <kleptog@svana.org>
http://svana.org/kleptog/
> It would be nice if someone came up with a certification system that
> actually separated those who can barely regurgitate what they crammed over
> the last few weeks from those who command secret ninja networking powers.