In my experience, ATT(SBC at that time) hit over its effective capacity (over 50% average utilization, and therefore no redundancy) around 2001. At least for clients I was working with, it was always evident that they didn't have enough capacity in any node to carry the traffic if they had a problem on any single upstream link. They also tended to manually handle routing decisions as opposed to letting the IGP handle it. Given the nature of the beast, I doubt that has changed much, and the anecdotal evidence posted here, most recently related to ATT/Cogent peering, bears that out. So, maybe from ATT's perspective the Internet (meaning their backbone) WILL be saturated by 2010. Since the Internet is a network of independent internets connected to each other, I'd like to know how Cicconi knows what the level of saturation of everyone else's backbone is, or their available dark capacity. I would think those are trade secrets that are closely guarded. It seems what we have here is ATT trying to create public hue and cry, so that the taxpayer will be compelled to pay for their required and overdue network upgrades, instead of themselves; or in order to get further regulatory relief in the name of investing in their infrastructure, as was done in the late '90s. Given their, and other's, track records with the subsidies and regulatory relief they were given in the late '90s, which they used to bankrupt the CLECs, and then passed the increased revenue onto shareholders, rather than investing in infrastructure, I'd be disinclined to give them what they want. The US lags the world in Broadband not because the FCC and PUCs hamstring the ILECs, but because of the disincentive for for-profit common stock companies with government granted monopolies to do much more than the bare minimum capital investment to keep operating costs low and competitors out of the market, while maximizing revenue from existing sunk cost. Would be competitors, on the other hand, have to make massive capital investments that require a long recovery period or high short-term prices, and are easily bankrupted by predatory pricing by the incumbents.

-----Original Message----- From: Sean Donelan [mailto:sean@donelan.com] Sent: Saturday, April 19, 2008 12:16 PM To: Scott Weeks Cc: nanog@nanog.org Subject: Re: [Nanog] ATT VP: Internet to hit capacity by 2010

On Fri, 18 Apr 2008, Scott Weeks wrote:

...

Does anybody know what the basis for Mr. Cicconi's claims were (if they even had a basis at all)?

Have there been an second reporting sources, or does anyone have a Youtube link of Mr. Cicconi's actual statement in context? So far there seems to only be a single reporter's account, echoed in the bloggerdome.

_______________________________________________ NANOG mailing list NANOG@nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

_______________________________________________ NANOG mailing list NANOG@nanog.org http://mailman.nanog.org/mailman/listinfo/nanog

Paul Wall wrote:

...

They also tended to manually handle routing decisions as opposed to letting the IGP handle it. Likewise, I'd be interested in implementation specifics of how a network of AT&T's caliber could implement backbone redundancy and TE with static routing.

atm-2, circuitzilla's dream machine. randy _______________________________________________ NANOG mailing list NANOG@nanog.org http://mailman.nanog.org/mailman/listinfo/nanog