[179860] in North American Network Operators' Group
Re: Rasberry pi - high density
daemon@ATHENA.MIT.EDU (Tim Raphael)
Sat May 9 01:48:58 2015
X-Original-To: nanog@nanog.org
From: Tim Raphael <raphael.timothy@gmail.com>
In-Reply-To: <3add94118911fb463b174ab24631657b@thefnf.org>
Date: Sat, 9 May 2015 13:48:49 +0800
To: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
The problem is, I can get more processing power and RAM out of two 10RU blad=
e chassis and only needing 64 10G ports...
32 x 256GB RAM per blade =3D 8.1TB
32 x 16 cores x 2.4GHz =3D 1,228GHz
(not based on current highest possible, just using reasonable specs)
Needing only 4 QFX5100s which will cost less than a populated 6513 and give l=
ower latency. Power, cooling and cost would be lower too.
RPi =3D 900MHz and 1GB RAM. So to equal the two chassis, you'll need:
1228 / 0.9 =3D 1364 Pis for compute (main performance aspect of a super comp=
uter) meaning double the physical space required compared to the chassis opt=
ion.
So yes, infeasible indeed.
Regards,
Tim Raphael
> On 9 May 2015, at 1:24 pm, charles@thefnf.org wrote:
>=20
>=20
>=20
> So I just crunched the numbers. How many pies could I cram in a rack?
>=20
> Check my numbers?
>=20
> 48U rack budget
> 6513 15U (48-15) =3D 33U remaining for pie
> 6513 max of 576 copper ports
>=20
> Pi dimensions:
>=20
> 3.37 l (5 front to back)
> 2.21 w (6 wide)
> 0.83 h
> 25 per U (rounding down for Ethernet cable space etc) =3D 825 pi
>=20
> Cable management and heat would probably kill this before it ever reached c=
ompletion, but lol...
>=20
>=20
>=20
