Re: heavly load system spec

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Gregg Jaskiewicz <gryzman(at)gmail(dot)com>
Cc: postgres performance list <pgsql-performance(at)postgresql(dot)org>
Subject: Re: heavly load system spec
Date: 2012-05-24 20:09:55
Message-ID: CA+TgmoYzt7gR3KWBBTM6RB0QUkBx++7ovz7qFUDnKLGFKaMCfg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Thu, Apr 5, 2012 at 11:39 AM, Gregg Jaskiewicz <gryzman(at)gmail(dot)com> wrote:
> I know this is a very general question. But if you guys had to specify
> system (could be one server or cluster), with sustainable transaction
> rate of 1.5M tps running postgresql, what configuration and hardware
> would you be looking for ?
> The transaction distribution there is 90% writes/updates and 10% reads.
> We're talking 64 linux, Intel/IBM system.
>
> I'm trying to see how that compares with Oracle system.

1.5 million is a lot of tps, especially if some of them are write
transactions. On trivial read-only transactions (primary key lookup
on fully cached table), using a 16-core, 64-thread IBM POWER7 box,
pgbench -M prepared -S -n -T 60 -c 64 -j 64:

tps = 455903.743918 (including connections establishing)
tps = 456012.871764 (excluding connections establishing)

That box isn't quite the fastest one I've seen, but it's close.

What hardware is Oracle running on?

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Stephen Frost 2012-05-24 21:04:05 Re: heavly load system spec
Previous Message Stephen Frost 2012-05-24 17:09:12 Re: High load average in 64-core server , no I/O wait and CPU is idle