Subject: Re: most bang for buck with ~ $20,000
From: Scott Marlowe <smarlowe@g2switchworks.com>
To: "Thomas F. O'Connell" <tfo@sitening.com>
Cc: "Joshua D.Drake" <jd@commandprompt.com>,
	Kenji Morishige <kenjim@juniper.net>, pgsql-performance@postgresql.org
In-Reply-To: <82C06632-148B-4489-90EB-BBB8A6BD1159@sitening.com>
References: <20060808204354.GN29379@juniper.net>
	<44D906E1.7010405@commandprompt.com>
	<9EE66C7F-0B04-4472-9EBD-CD567AC9EA1D@sitening.com>
	<44D90FFE.7010807@commandprompt.com>
	<82C06632-148B-4489-90EB-BBB8A6BD1159@sitening.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Message-Id: <1155136527.20252.90.camel@state.g2switchworks.com>
Mime-Version: 1.0
Date: Wed, 09 Aug 2006 10:15:27 -0500

On Tue, 2006-08-08 at 17:53, Thomas F. O'Connell wrote:
> On Aug 8, 2006, at 5:28 PM, Joshua D. Drake wrote:
> 
> > Thomas F. O'Connell wrote:
> >> On Aug 8, 2006, at 4:49 PM, Joshua D. Drake wrote:
> >>>> I am considering a setup such as this:
> >>>>   - At least dual cpu (possibly with 2 cores each)
> >>>>   - 4GB of RAM
> >>>>   - 2 disk RAID 1 array for root disk
> >>>>   - 4 disk RAID 1+0 array for PGDATA
> >>>>   - 2 disk RAID 1 array for pg_xlog
> >>>> Does anyone know a vendor that might be able provide such setup?
> >> Wouldn't it be preferable to put WAL on a multi-disk RAID 10 if  
> >> you had the opportunity? This gives you the redundancy of RAID 1  
> >> but approaches the performance of RAID 0, especially as you add  
> >> disks to the array. In benchmarking, I've seen consistent success  
> >> with this approach.
> >
> > WALL is written in order so RAID 1 is usually fine. We also don't  
> > need journaling for WAL so the speed is even faster.
> 
> In which case, which is theoretically better (since I don't have a  
> convenient test bed at the moment) for WAL in a write-heavy  
> environment? More disks in a RAID 10 (which should theoretically  
> improve write throughput in general, to a point) or a 2-disk RAID 1?  
> Does it become a price/performance question, or is there virtually no  
> benefit to throwing more disks at RAID 10 for WAL if you turn off  
> journaling on the filesystem?

Actually, the BIGGEST win comes when you've got battery backed cache on
your RAID controller.  In fact, I'd spend money on a separate RAID
controller for xlog with its own cache hitting a simple mirror set
before I'd spring for more drives on pg_xlog.  The battery backed cache
on the pg_xlog likely wouldn't need to be big, just there and set to
write-back.

Then put all the rest of your cash into disks on a big RAID 10 config,
and as big of a battery backed cache as you can afford for it and memory
for the machine.