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

Re: Large database help

From: xbdelacour(at)yahoo(dot)com
To: Ragnar Kjørstad <postgres(at)ragnark(dot)vestdata(dot)no>
Cc: pgsql-admin(at)postgresql(dot)org
Subject: Re: Large database help
Date: 2001-04-22 22:52:26
Message-ID: 5.0.2.1.0.20010422184417.01972e70@pop.mail.yahoo.com (view raw or flat)
Thread:
Lists: pgsql-admin
I'm spawning 6 backends to query the data. top lists 6 postmaster processes 
working, and therefore the idle time should hit 0% easily. Also, the hard 
drive light goes nuts when I'm running this.

Here is the pertinent information from top. To be clear, I'm NOT spawning a 
new postmaster per chunk. These same six processes are alive and kicking 
for over 4 minutes.

I hope the formatting works ok.

   PID USER     PRI  NI  SIZE  RSS SHARE STAT  LIB %CPU %MEM   TIME COMMAND
  2379 mg        13   0 14476  14M 12908 S       0  5.8  2.7   0:01 postmaster
  2380 mg        10   0 14436  14M 12868 S       0  2.9  2.7   0:01 postmaster
  2381 mg        10   0 13572  13M 12012 S       0  2.9  2.6   0:00 postmaster
  2377 mg        10   0 13640  13M 12072 S       0  2.4  2.6   0:01 postmaster
  2378 mg        11   0 14476  14M 12908 S       0  2.4  2.7   0:01 postmaster
  2376 mg         8   0 13556  13M 11984 S       0  1.9  2.6   0:00 postmaster

27 processes: 24 sleeping, 3 running, 0 zombie, 0 stopped
CPU states: 16.3% user,  3.8% system,  0.0% nice, 79.8% idle
Mem:  517292K av, 508400K used,   8892K free,  94444K shrd, 197224K buff
Swap:  65988K av,      0K used,  65988K free                160740K cached

-Xavier

At 12:46 AM 4/23/01 +0200, Ragnar Kjørstad wrote:
>On Sun, Apr 22, 2001 at 05:12:20PM -0400, xbdelacour(at)yahoo(dot)com wrote:
> > My problem is this: during the query process the hard drive is being 
> tagged
> > excessively, while the cpu's are idling at 50% (numbers from Linux 
> command:
> > top), and this is bringing down the speed pretty dramatically since the
> > process is waiting on the hard disk. How do I get the database to be
> > completely resident in memory such that selects don't cause hdd activity?
> > How do I pin how exactly why the hard disk is being accessed?
>
>50% idle on a 2 CPU system mean one CPU is busy and one is idle - as
>expected if you run a single CPU-intensive job.
>
> > I am setting 'echo 402653184 >/proc/sys/kernel/shmmax', which is being
> > reflected in top. I also specify '-B 48000' when starting postmaster. My
> > test DB is only 86MB, so in theory the disk has no business being active
> > once the data is read into memory unless I perform a write operation.. 
> What
> > am I missing?
>
>What does top say?
>How much memory do each of your postgres processes take, and how much of
>it is shared?
>Is there any free memory?
>
>
>--
>Ragnar Kjørstad


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


In response to

Responses

pgsql-admin by date

Next:From: Randall PerryDate: 2001-04-22 23:54:25
Subject: Connecting via perl gives "root" does not exist
Previous:From: Ragnar KjørstadDate: 2001-04-22 22:46:46
Subject: Re: Large database help

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