From: Evgeniy Shishkin <itparanoia@gmail.com>
To: Graeme B. Bell <grb@skogoglandskap.no>
Cc: Andrea Suisani <sickpig@opinioni.net>
Cc: mfatticcioni@mbigroup.it <mfatticcioni@mbigroup.it>
Cc: pgsql-performance@postgresql.org <pgsql-performance@postgresql.org>
Subject: Re: Tuning the configuration
Date: Tue, 16 Dec 2014 19:37:28 +0300
Message-ID: <63042BAB-3B28-441F-B819-B2DDC8FB24EE@gmail.com> (raw)
In-Reply-To: <BB91D876-5FE8-4938-9FEF-8984861E1536@skogoglandskap.no>
References: <548823F8.4070003@mbigroup.it>
<548987D7.7040304@opinioni.net>
<6B084BAC-82C6-410A-88D4-9CDC01DA8C09@gmail.com>
<BB91D876-5FE8-4938-9FEF-8984861E1536@skogoglandskap.no>
List-Unsubscribe: <mailto:majordomo@postgresql.org?body=unsub%20pgsql-performance>
> On 16 Dec 2014, at 14:51, Graeme B. Bell <grb@skogoglandskap.no> wrote:
>
>>
>> I don't understand the logic behind using drives,
>> which are best for random io, for sequent io workloads.
>
> Because they are also best for sequential IO. I get 1.3-1.4GB/second from 4 SSDs in RAID or >500MB/s for single disk systems, even with cheap models.
> Are you getting more than that from high-end spinning rust?
I better use ssd for random iops when database doesn't fit in ram.
For wal logs i use raid with bbu cache and couple of sas drives.
--
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: itparanoia@gmail.com, grb@skogoglandskap.no, sickpig@opinioni.net, mfatticcioni@mbigroup.it
Subject: Re: Tuning the configuration
In-Reply-To: <63042BAB-3B28-441F-B819-B2DDC8FB24EE@gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox