Subject: Re: PostgreSQL
To: Greg 'groggy' Lehey <grog@NetBSD.org>
From: Curt Sampson <email@example.com>
Date: 02/03/2006 14:32:57
On Fri, 3 Feb 2006, Greg 'groggy' Lehey wrote:
> On Thursday, 2 February 2006 at 15:37:29 +0000, Patrick Welche wrote:
>> On Wed, Feb 01, 2006 at 06:58:04PM -0800, Alfred Perlstein wrote:
>>> * Neil Conway <firstname.lastname@example.org> [060201 18:49] wrote:
>>>> On Thu, 2006-02-02 at 00:49 +0000, email@example.com wrote:
>>> But threads aren't always the best thing out there, the code gets
>>> a lot more complicated and crash prone and you wind up sometimes
>>> getting screwed because third party code isn't thread safe...
>> Isn't your last paragraph the reason PostgreSQL sticks to processes?
> I'd guess no. It might be a good reason to do so, but the real reason
> is what I stated in a previous message: it was the only option at the
> time. And "if it works, don't fix it".
Well, I'm not a PostgreSQL developer, but I reckon that this is not it.
Looking at it from my point of view, I think there's also the, "what
does it get you" factor.
Typically, DBMSes under high load have a lot of different queries going
on a the same time. For that, threading doesn't help at all over just
having a separate process doing each query.
I can think of situations where having one query running across multiple
processors would be an advantage, but they seem pathological to me.
Curt Sampson <firstname.lastname@example.org> +81 90 7737 2974
The power of accurate observation is commonly called cynicism
by those who have not got it. --George Bernard Shaw