Re: Setting "nice" values - Mailing list pgsql-performance

From Madison Kelly
Subject Re: Setting "nice" values
Date
Msg-id 454F3D23.6040003@alteeve.com
Whole thread Raw
In response to Re: Setting "nice" values  (Tobias Brox <tobias@nordicbet.com>)
Responses Re: Setting "nice" values
List pgsql-performance
Tobias Brox wrote:
> [Madison Kelly - Mon at 08:10:12AM -0500]
>> to run, which puts it into your "drawback" section. The server in
>> question is also almost under load of some sort, too.
>>
>> A great tip and one I am sure to make use of later, thanks!
>
> I must have been sleepy, listing up "cons" vs "drawbacks" ;-)

:) I noticed but figured what you meant (I certainly do similar flubs!).

> Anyway, the central question is not the size of the job, but the size of
> the transactions within the job - if the job consists of many
> transactions, "my" test can be run before every transaction.  Having
> transactions lasting for hours is a very bad thing to do, anyway.

Ah, sorry, long single queries is what you meant. I have inherited this
code so I am not sure how long a given query takes, though they do use a
lot of joins and such, so I suspect it isn't quick; indexes aside. When
I get some time (and get the backup server running) I plan to play with
this. Currently the DB is on a production server so I am hesitant to
poke around just now. Once I get the backup server though, I will play
with your suggestions. I am quite curious to see how it will work out.

Thanks again!

Madi


pgsql-performance by date:

Previous
From: Tobias Brox
Date:
Subject: Re: Setting "nice" values
Next
From: Hannes Dorbath
Date:
Subject: Yet another question on LIMIT performance :/