[Beowulf] torus versus (fat) tree topologies
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Don Holmgren djholm at fnal.govFri Nov 12 17:00:11 PST 2004
- Previous message: [Beowulf] torus versus (fat) tree topologies
- Next message: [Beowulf] Bioinformatics/Beowulf applications
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Fri, 12 Nov 2004, Patrick Geoffray wrote:
> Hi Håkon,
>
> Håkon Bugge wrote:
> > deployment times and more complicated node replacement. In larger
> > systems though, cabling of centralized switched tends to require very
> > _long_ cables, something you do not need using tori topologies.
>
> How do you close your Torus without long cables ? Unless you stack your
> nodes in a circle, you will need long cables.
>
> Patrick
Hi Patrick -
Interleave the cables. Suppose you have 10 nodes. The first way people
think about cabling a torus of 10 nodes is:
1 -- 2 -- 3 -- 4 -- 5 -- 6 -- 7 -- 8 -- 9 -- 10
| |
----------------------------------------------
In this configuration, yes, there's a long cable from 1 to 10 to close
the loop. The shorthand way of writing this is:
1-2-3-4-5-6-7-8-9-10-1
Instead, interleave the cables connecting every other node:
1-3-5-7-9-10-8-6-4-2-1
The longest cable in this scheme is twice the distance between each
adjacent pair of nodes.
Don
- Previous message: [Beowulf] torus versus (fat) tree topologies
- Next message: [Beowulf] Bioinformatics/Beowulf applications
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
