Skip site navigation (1) Skip section navigation (2)

Peripheral Links

Header And Logo

PostgreSQL
| The world's most advanced open source database.

Site Navigation

Search for
  Advanced Search

Re: New to PostgreSQL, performance considerations


  • From: Cosimo Streppone <cosimo(at)streppone(dot)it>
  • To: Postgresql Performance list <pgsql-performance(at)postgresql(dot)org>
  • Cc: Michael Stone <mstone+postgres(at)mathom(dot)us>
  • Subject: Re: New to PostgreSQL, performance considerations
  • Date: Wed, 13 Dec 2006 17:11:59 +0100
  • Message-id: <4580264F(dot)8030109(at)streppone(dot)it>

Michael Stone wrote:

On Tue, Dec 12, 2006 at 01:42:06PM +0100, Cosimo Streppone wrote:
"-O0" ~ 957 tps
"-O1 -mcpu=pentium4 -mtune=pentium4" ~ 1186 tps
"-O2 -mcpu=pentium4 -mtune=pentium4" ~ 1229 tps
"-O3 -mcpu=pentium4 -mtune=pentium4" ~ 1257 tps
"-O6 -mcpu=pentium4 -mtune=pentium4" ~ 1254 tps

I'm curious now to get the same tests run with
a custom-cflags-compiled glibc.

I'd be curious to see -O2 with and without the arch-specific flags, since that's mostly what the discussion is about.

I run the same tests only for:

1) '-O2'
2) '-O2 -march=pentium4 -mtune=pentium4 -mcpu=pentium4'
   (so no more doubts here, and thanks for gcc hints :-)

and I obtained respectively an average of 1238 (plain -O2)
vs. 1229 tps on 9 runs.
Disk subsystem is a standard desktop SATA, no more than that.

I tried also recompiling *only* pgbench with various options, but as
I expected (and hoped) nothing changed.

Interesting, eh?

--
Cosimo




Home | Main Index | Thread Index

Privacy Policy | PostgreSQL Archives hosted by Command Prompt, Inc. | Designed by tinysofa
Copyright © 1996 – 2008 PostgreSQL Global Development Group