Re: INDEX_MAX_KEYS to 64?

From: Rod Taylor <rbt(at)rbt(dot)ca>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: PostgreSQL Development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: INDEX_MAX_KEYS to 64?
Date: 2003-07-01 01:40:28
Message-ID: 1057023628.28657.130.camel@jester
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, 2003-07-01 at 01:25, Tom Lane wrote:
> Rod Taylor <rbt(at)rbt(dot)ca> writes:
> > I just noticed that the OSDL benchmarks for PostgreSQL appear to require
> > PostgreSQL be compiled with INDEX_MAX_KEYS as 64 rather than the default
> > of 32.
>
> Which one? I've been testing dbt3 here and not seen that.

It's in the directions for dbt2, though I haven't found the specific
function(s) that requires it at this point -- can't quite run it on my
system yet for various reasons.

> > Any chance the default could be bumped for the 7.4 release?
>
> Not without evidence that it doesn't cause performance penalties.
> ISTM we have been through this discussion recently, and concluded
> that 32 was the place to set it.

Yes, I was digging through that discussion. The test used shows a 4%
difference between 32 and 64.

do 100 times
select 2+2+2+2+2+2+ ... iterated 9901 times

#define INDEX_MAX_KEYS 16, 32, 64, & 128
#define FUNC_MAX_ARGS INDEX_MAX_KEYS
make all
make install
initdb

The results were as follows:
INDEX_MAX_KEYS 16 32 64 128
-----+-------+------+--------
Time in seconds 48 49 51 55

--
Rod Taylor <rbt(at)rbt(dot)ca>

PGP Key: http://www.rbt.ca/rbtpub.asc

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Joe Conway 2003-07-01 02:01:52 Re: INDEX_MAX_KEYS to 64?
Previous Message Tom Lane 2003-07-01 01:25:25 Re: INDEX_MAX_KEYS to 64?