Skip site navigation (1) Skip section navigation (2)

Re: average query performance measuring

From: Stephen Frost <sfrost(at)snowman(dot)net>
To: Karl Denninger <karl(at)denninger(dot)net>
Cc: Rick Otten <rotten(at)manta(dot)com>,"pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org>
Subject: Re: average query performance measuring
Date: 2012-08-21 20:27:02
Message-ID: 20120821202702.GO1267@tamriel.snowman.net (view raw or flat)
Thread:
Lists: pgsql-performance
Karl,

* Karl Denninger (karl(at)denninger(dot)net) wrote:
> That looks EXTREMELY useful and I'm looking forward to checking it out
> in 9.2; I have asked a similar question about profiling actual queries
> in the past and basically it came down to "turn on explain or run a
> separate explain yourself since the app knows what's similar and what's
> not", which of course has hideous performance implications (as the query
> basically executes twice.)

Just to clarify one thing- if your application is currently using
prepared queries for everything, you can probably use the existing
contrib module.  The difference is that, with 9.2, it'll actually do
normalization of non-PREPARED queries and will include some additional
statistics and information.

	Thanks,

		Stephen

In response to

pgsql-performance by date

Next:From: Craig JamesDate: 2012-08-21 20:59:06
Subject: Re: Does setval(nextval()+N) generate unique blocks of IDs?
Previous:From: Scott MarloweDate: 2012-08-21 20:04:59
Subject: Re: Does setval(nextval()+N) generate unique blocks of IDs?

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group