Channel-bonding - choice of switch?
jlargent at imagelinks.com
Thu Jul 12 08:29:38 PDT 2001
Josip Loncaric wrote:
> Jakob Østergaard wrote:
> > You can buy separate switches, unmanaged dirt cheap ones.
> Sounds reasonable, and if we did use channel bonding, that's what we'd
> do. This "separate switches" solution results in two identical but
> physicaly separate LANs, which means that you also need a bridge machine
Don't you have this anyway?
It's my understanding that channel bonding is an all or none
situation. You can't have a nonbonded workstation going to one
of the vlans, and then use that workstation to talk to the bonded
cluster nodes. You lose half a your packets coming back from the
bonded node, so those are retransmitted and half of those are lost
and so on, until eventually you will get all the packets but with
many many retranmissions.
> between channel-bonded nodes and normal machines. The bridge machine
> can be a bottleneck for applications that need high bandwidth to normal
> machines (data servers, video displays).
> Dr. Josip Loncaric, Research Fellow mailto:josip at icase.edu
> ICASE, Mail Stop 132C PGP key at http://www.icase.edu./~josip/
> NASA Langley Research Center mailto:j.loncaric at larc.nasa.gov
> Hampton, VA 23681-2199, USA Tel. +1 757 864-2192 Fax +1 757 864-6134
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
Jeff Largent ImageLinks, Inc.
Sr System Admin Melbourne, Fl 32935
(321) 253-0011 fax:(321) 253-5559
More information about the Beowulf