Telecom Collapse?

Alexander Harrowell a.harrowell at gmail.com
Thu Dec 4 18:16:38 CST 2008


On Thu, Dec 4, 2008 at 7:18 PM, Michael Thomas <mike at mtcc.com> wrote:

>   We haven't really had a major catastrophe where we've been totally
>  dependent on IP yet, AFIAK. Maybe all of the qos, call gapping and
>  the rest of the stuff the TDM networks do to deal with disasters
>  will be left in the dustbin of Moore's Law, but maybe they won't. One
>  thing is certain: we'll definitely find out one day, and it's not
>  likely to be from a position of having taken the precautions,
>  congratulating ourselves IMO.


 The only disaster I experienced which affected telecoms was the July, 2005
terrorist attack on London. Although the infrastructure wasn't affected,
there were significant load challenges for the GSM nets especially. It was
widely assumed by the unclued that either one or two GSM operators failed
under peak load; by the clued that the Access Overload Control process,
analogous to the PSTN's Government Telephone Preference Scheme, had been
initiated to deal with the peak load.

In fact, it turned out much later that AOC had indeed been declared, but
unnecessarily, and against the decision of the lead agency dealing with the
emergency. The Metropolitan Police didn't request it, but the (smaller) City
of London force did, although the network in question was coping - the
entire outage was caused by mismanaging the TDM call-gapping and QoS
features.

Both the Internet, and our corporate VoIP system including its peering with
the wider PSTN, worked throughout.



More information about the NANOG mailing list