From: Andy Ballingall <andy_ballingall@bigfoot.com>
To: J. Andrew Rogers <jrogers@neopolitan.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Working on huge RAM based datasets
Date: Sat, 10 Jul 2004 13:25:37 +0100
Message-ID: <021101c46679$01ef4330$0300a8c0@lappy> (raw)
References: <20040708101913.GA15871@uio.no>
<00ac01c46516$3ec859e0$0300a8c0@lappy>
<40EDF7DE.8080900@familyhealth.com.au>
<011301c46597$15d145c0$0300a8c0@lappy>
<1089409248.4524.13.camel@vulture.corp.neopolitan.com>
Oops - sorry - I confused my numbers. The opteron machine in mind *only* has
up to 64GB of RAM (e.g. HP DL585) - here's the datapage:
http://h18004.www1.hp.com/products/servers/proliantdl585/index.html
Still - with *just* 64GB of RAM, that would comfortably provide for the type
of scenario I envisage. Is that still enough for your app?
The 256GB number came from something I read saying that the current crop of
64 bit chips will allow up to 256GB of RAM in principle, so it is just a
matter of time before the memory limit shoots up on these simple products.
If you are prepared to pay a bit more, already there are some big memory
options on linux:
E.g. you can have up to 192GB in an SGI Altix 350:
http://www.sgi.com/servers/altix/downloads/altix350_at_a_glance.pdf
Or up to 4 terabytes in it's bigger brother the Altix 3000 - but that's
getting a bit esoteric.
http://www.sgi.com/servers/altix/
(This won lots of awards recently)
The nice thing about the two things above is that they run linux in a single
address space NUMA setup, and in theory you can just bolt on more CPUs and
more RAM as your needs grow.
Thanks,
Andy
----- Original Message -----
From: "J. Andrew Rogers" <jrogers@neopolitan.com>
To: "Andy Ballingall" <andy_ballingall@bigfoot.com>
Sent: Friday, July 09, 2004 10:40 PM
Subject: Re: [PERFORM] Working on huge RAM based datasets
> On Fri, 2004-07-09 at 02:28, Andy Ballingall wrote:
> > After all, we're now seeing the first wave of 'reasonably priced' 64 bit
> > servers supported by a proper 64 bit OS (e.g. linux). HP are selling a 4
> > Opteron server which can take 256GB of RAM, and that starts at $10000
(ok -
> > they don't give you that much RAM for that price - not yet, anyway!)
>
>
> Which server is this?! They are selling an Opteron system that can hold
> 256 GB of RAM?
>
> I looked on their site, and couldn't find anything like that. I run
> some MASSIVE memory codes that don't need a lot of CPU, and if such a
> box existed, I'd be very interested.
>
> cheers,
>
> j. andrew rogers
>
>
>
----- Original Message -----
From: "J. Andrew Rogers" <jrogers@neopolitan.com>
To: "Andy Ballingall" <andy_ballingall@bigfoot.com>
Sent: Friday, July 09, 2004 10:40 PM
Subject: Re: [PERFORM] Working on huge RAM based datasets
> On Fri, 2004-07-09 at 02:28, Andy Ballingall wrote:
> > After all, we're now seeing the first wave of 'reasonably priced' 64 bit
> > servers supported by a proper 64 bit OS (e.g. linux). HP are selling a 4
> > Opteron server which can take 256GB of RAM, and that starts at $10000
(ok -
> > they don't give you that much RAM for that price - not yet, anyway!)
>
>
> Which server is this?! They are selling an Opteron system that can hold
> 256 GB of RAM?
>
> I looked on their site, and couldn't find anything like that. I run
> some MASSIVE memory codes that don't need a lot of CPU, and if such a
> box existed, I'd be very interested.
>
> cheers,
>
> j. andrew rogers
>
>
>
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: andy_ballingall@bigfoot.com, jrogers@neopolitan.com
Subject: Re: Working on huge RAM based datasets
In-Reply-To: <021101c46679$01ef4330$0300a8c0@lappy>
* 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