Re: Hash Join vs Nested Loops in 7.2.1 ... - Mailing list pgsql-general

From Ed Loehr
Subject Re: Hash Join vs Nested Loops in 7.2.1 ...
Date
Msg-id 3CB27DF3.8010108@bluepolka.net
Whole thread Raw
In response to Hash Join vs Nested Loops in 7.2.1 ...  (Ed Loehr <pggeneral@bluepolka.net>)
Responses Re: Hash Join vs Nested Loops in 7.2.1 ...
List pgsql-general
Tom Lane wrote:

>
>>Second, the same query sometimes takes 10-50 seconds shortly after
>>possibly a dump or other high-data-volume queries are executed, after
>>which it then returns to 1 second execution time.  Getting crowded out
>>of shared memory?
>
> Sounds like it.  What shared-buffers setting are you using?  How much
> RAM in the box?


shared_buffers = 256
max_fsm_relations = 500
(defaults for the rest)

RAM:  2.4GB, maybe?  Not that familiar with HPUX mem setup...

(OS: HP-UX B.11.00 U 9000/800)
$ swapinfo -mt
              Mb      Mb      Mb   PCT  START/      Mb
TYPE      AVAIL    USED    FREE  USED   LIMIT RESERVE  PRI  NAME
dev        2048     147    1901    7%       0       -    1  /dev/vg00/lvol2
reserve       -     312    -312
memory      369     351      18   95%
total      2417     810    1607   34%       -       0    -

> AFAICS the planner will do this one just fine
> without any thumb on the scales.


How to I find the thumb?

Ed


pgsql-general by date:

Previous
From: Tom Lane
Date:
Subject: Re: Hash Join vs Nested Loops in 7.2.1 ...
Next
From: Jean-Michel POURE
Date:
Subject: Re: Porting from MySql - meta data issues