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

Re: [PERFORM] Interesting incosistent query timing

From: Ernest E Vogelsinger <ernest(at)vogelsinger(dot)at>
To: "Nikolaus Dilger" <nikolaus(at)dilger(dot)cc>
Cc: pgsql-performance(at)postgresql(dot)org, pgsql-general(at)postgresql(dot)org
Subject: Re: [PERFORM] Interesting incosistent query timing
Date: 2003-06-17 02:54:56
Message-ID: 5.1.1.6.2.20030617044100.02dd1c70@mail.vogelsinger.at (view raw or flat)
Thread:
Lists: pgsql-generalpgsql-performance
At 04:20 17.06.2003, Nikolaus Dilger said:
--------------------[snip]--------------------
>My guess is that the second execution of the query is
>shorter since the data blocks are cached in memory. 
>When you modify the data then it needs to be read again
>from disk which is much slower than from memory.  The
>short execution after restarting PostgreSQL seems to
>indicate that your data is cached in the Linux buffer
>cache. 
>
>The only strange thing seems to be that you have so few
>rows.  Are you getting the data from a remote machine? 
>How many bytes does a single row have?  Are they really
>large???
--------------------[snip]-------------------- 

What exactly do you mean? This table is quite filled (2.3 million rows),
but the query results are correct.


-- 
   >O     Ernest E. Vogelsinger
   (\)    ICQ #13394035
    ^     http://www.vogelsinger.at/



pgsql-performance by date

Next:From: Dawn HollingsworthDate: 2003-06-17 06:46:48
Subject: Re: Postgres Connections Requiring Large Amounts of Memory
Previous:From: Nikolaus DilgerDate: 2003-06-17 02:20:57
Subject: Re: [PERFORM] Interesting incosistent query timing

pgsql-general by date

Next:From: Tom LaneDate: 2003-06-17 03:32:05
Subject: Re: order of nested loop
Previous:From: Nikolaus DilgerDate: 2003-06-17 02:20:57
Subject: Re: [PERFORM] Interesting incosistent query timing

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