| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Ken Williams <ken(at)mathforum(dot)org> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Optimizing further |
| Date: | 2002-06-14 03:03:14 |
| Message-ID: | 2225.1024023794@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Ken Williams <ken(at)mathforum(dot)org> writes:
> On Friday, June 14, 2002, at 12:35 PM, Tom Lane wrote:
>> That sounds like a caching issue. How much RAM do you have,
>> what have you got shared_buffers set to, etc?
> 'top' reports 1024M of real memory, with 819M virtual. shared_buffers
> is currently set to 64.
Hm. If there were nothing else going on in your system, I'd expect
the kernel to buffer disk pages more or less indefinitely. Do you
have other processes running that might be sucking memory away from
kernel disk buffers?
Most people think that shared_buffers=64 is a ridiculously low
default, and that something in the low thousands is more realistic for
a production database. However I'm not sure that bumping that up will
help for your particular issue. I'm guessing that your machine is
under memory pressure and that what you really need to do is buy more
RAM. Does 'top' tell you anything about the amount of RAM being used
for kernel disk buffers?
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Doug Fields | 2002-06-14 03:08:16 | Re: Optimizing further |
| Previous Message | Ken Williams | 2002-06-14 02:45:50 | Re: Optimizing further |