DC fiber cut: field report

Sean M. Doran smd at clock.org
Wed Jul 23 13:13:54 UTC 1997


Vadim Antonov <avg at pluris.com> writes:

> Yep.  I would say that telcos have rather strong incentive to
> keep PDH in place.  It may actually be a good deal for IP people --
> you can get more capacity at a lower price, and handle redundancy
> at level 3 (hey, interior re-routing is fast).

You've been retired for too long when you start believing
that telco price == good deal for any package at all.

Fortunately, traditional PNOs are facing real competition
in various parts of the world already, with all hell
breaking loose in these parts Real Soon Now.

> Of course, to get advantage of that practice one may want to
> work closely with telco to understand their modes of failure,
> and write a language protecting topology of his network
> overlay.

Well, maybe if you *were* the telco this *might* work a la
FMS. :)  

In practice, though, I would want to have redundancy there
in a simple way, namely through SDH or SONET.   I believe
this is what lots of carriers are negotiating amongst
themselves (each not trusting the others to actually
provide real PDH protection redundancy properly), which
may be of some follow-on benefit to the retail market.

Also, I think you will find plenty of people here who are
so fed up with backhoe fade and broken PDH protection
switching that they will pay good money for capacity
provisioned over SONET rings.

Ironically, facilities-wise, SONET is much cheaper than
PDH support for protection-switching.

> >You don't need an additional MUX, you don't need extra DAXCs,
> >and you don't need to scurry around reprogramming lots of equipment
> >when someone wrecks your fibre.
> 
> It's technical staff who's doing all the scurrying.  Sales people
> already got their comission and are quite happy.

While this is healthy cynicism about PNOs, there is the
potential pain of heavy downtime penalty clauses that I
hope all major ISPs and other purchasers have started
getting written into their capacity contracts when they're
paying anything close to normal prices.

> My prediction is that redundancy of telco data services won't improve
> at all until they find a way to reconcile need for redundancy and
> desire to sell all the capacity they got. 

I don't agree.  In fact, there are extant counterexamples.
Unreliable service costs business, and SONET/SDH rings
elminate most of the most common causes of unreliable
transmissions facilities.  

I would hope that Worldcom's PNO side will lose alot of
revenues, in the face of the recent cuts, which amply
demonstrated their non-deployment of SONET rings along the
fibre paths in question.

They deserve it, too.  Shame on them.

	Sean.



More information about the NANOG mailing list