RE: Requirements for IP Multihoming Architectures

[Antonio Querubin <tony@lava.net>]

On Tue, 20 Mar 2001, Vijay Gill wrote:

> On Tue, 20 Mar 2001, Elwyn Davies wrote:
>
> > 2. Redundancy is illusory without getting into the 'shared fate'
> > considerations which are currently exercising the IPO working group.   If
> > your multiple connections end up running in a single duct that gets back
> > hoed all your hard work goes for nothing.
>
> There are multiple failure modes that can be protected against. If you are
> connected via a single conduit run, then the back hoe fade protection of
> dual homing is not going to happen. However, you are still protected
> against failure at the IP level, including AS wide failures of a provider.
>
> Now is it worth protecting against an AS/router/card failure when you
> cannot protect against a backhoe?  from my time in the enterprise market,
> I would lean towards - yes.

I would concur.  I can't count the number of times we've been affected by
line interface card degradation (or complete failure) in some telco or
upstream providers' equipment.  But backhoe type outages are actually
quite rare.

However, the latter gets more attention I think because the estimated time
to repair can be uncomfortably ambiguous and usually long.  A line card
failure ETTR can more easily be pinned down to whether spare cards are
available in-house (usually not) or how fast a replacement can be shipped.