Re: [HACKERS] Big IN() clauses etc : feature proposal - Mailing list pgsql-performance

From Jim C. Nasby
Subject Re: [HACKERS] Big IN() clauses etc : feature proposal
Date
Msg-id 20060511200028.GM99570@pervasive.com
Whole thread Raw
In response to Re: [HACKERS] Big IN() clauses etc : feature proposal  (Martijn van Oosterhout <kleptog@svana.org>)
List pgsql-performance
On Thu, May 11, 2006 at 08:43:46PM +0200, Martijn van Oosterhout wrote:
> On Thu, May 11, 2006 at 11:35:34AM -0400, Greg Stark wrote:
> > I can say that I've seen plenty of instances where the ability to create
> > temporary tables very quickly with no overhead over the original query would
> > be useful.
>
> I wonder if this requires what the standard refers to as a global
> temporary table. As I read it (which may be wrong, I find the language
> obtuse), a global temporary table is a temporary table whose structure
> is predefined. So, you'd define it once, updating the catalog only once
> but still get a table that is emptied each startup.
>
> Ofcourse, it may not be what the standard means, but it still seems
> like a useful idea, to cut down on schema bloat.

IIRC that's the exact syntax Oracle uses:

CREATE GLOBAL TEMPORARY TABLE ...

I always found it a bit odd, since it always seemed to me like a global
temporary table would be one that every backend could read... something
akin to a real table that doesn't worry about fsync or any of that (and
is potentially not backed on disk at all).
--
Jim C. Nasby, Sr. Engineering Consultant      jnasby@pervasive.com
Pervasive Software      http://pervasive.com    work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf       cell: 512-569-9461

pgsql-performance by date:

Previous
From: "Jim C. Nasby"
Date:
Subject: Re: [HACKERS] Big IN() clauses etc : feature proposal
Next
From: "Jim C. Nasby"
Date:
Subject: Re: [HACKERS] Big IN() clauses etc : feature proposal