Our Networks; Not Ready for a Disaster
A public-private initiative, one that is more than talk, will be necessary to create the network services resiliency we already need but do not have
This year's natural disasters in the Northeast and Mid-Atlantic regions, where a major portion of the U.S. population lives, have created more emergencies than anticipated. They have demonstrated that the carriers, providers and the FCC lag in delivering more resilient networks.
Disasters in My Life
I experienced the earthquake in August 2011, the Derecho storm in June 2012 and Sandy in October 2012. The earthquake was scary but not a disruption in my area, but it did affect southern Virginia. The Derecho storm and Sandy did cause problems. However, we had fewer communications and power interruptions than might be expected because of Sandy. Most of the trees that could disrupt communications and power had already fallen down during the Derecho storms, so fewer trees fell during Sandy, causing fewer outages.
The article posted at Knowledge @Wharton, "Disconnected? The Perils of Digital Interdependence" revived my interest in communications during and after emergencies. Like many discussions of communications failures, this article discusses the need to preserve communications for first responders and public safety. Many of these agencies use wireless services to perform their jobs, so wireless loss equals ineffective or late emergency response. The Wharton article also discusses communications loss encountered by consumers and businesses.
Some of this communications loss was due to power loss. You can't access broadband IP services if your modem and router are down. As it turns out, the mobile services also took hits from Sandy. However, some of the wireless loss was due a lack of long-term power backup for the wireless towers.
Closing the PSTN?
I grew up and worked with the gold-plated PSTN. This was effectively mandated by the FCC. Now AT&T wants to retire the resilient TDM PSTN, as discussed in its "AT&T Petition to Launch a Proceeding Concerning the TDM-to-IP Transition". If the petition is successful, then AT&T will have fewer responsibilities. One reason AT&T wants to move away from the PSTN is that the PSTN is not as profitable as their other services.
The argument that the landlines are decreasing in use is correct. But the decrease is a lot slower than AT&T would have you believe. A phase-out of the TDM network will eventually happen. But replacing it with an all-IP network that does not produce the same resiliency as the PSTN will only lead to more expensive consumer and business losses, besides the impact on first responders using wireless communications.
My NoJitter article "PSTN Closure, the End of POTS, the Challenges" discusses the termination of the PSTN and the expected ramifications. I don't think we are ready to abandon the PSTN yet. The loss of communications during Sandy did not prove that mobile networks were any more reliable than landline and VoIP cable networks.
The FCC Responds to Communications Disasters
In response to the communications problems encountered with these emergencies, the FCC has announced hearings on new challenges to network resiliency. This will certainly open a lot of questions and create a number of strategic plans. But I am not sure it will go anywhere. There is no financial incentive for carriers and providers to build the rock-solid networks that could withstand events like Sandy in the future. The carriers and providers will shout that this resiliency level will raise their service costs. If the FCC tries to mandate resiliency improvements, there will be pushback by the lobbyists. The federal government, with its financial limitations, cannot afford to pay for the improvements. Maybe we should have a communications continuity tax like the USF to fund the resiliency improvements.
The point I made in my previous blog, "Five Nines and Acts of God", was that "The Derecho storm and outages in Virginia demonstrate the need for enterprises to understand the carriers' systems and potential shortcomings." This article covered the poor Verizon support for 911 calls during and for 3-1/2 days after that storm passed. 911 in northern Virginia was a mess because Verizon did not have adequate backup facilities and lost two major central offices for a time. This was mainly due to inadequate power backup capabilities. Even when my FiOS Internet and TV service resumed, I could not make or receive phone calls on either my FiOS cable connection or my POTS landline for days. In portions of Verizon's service areas, these problems hampered first responders' ability to deliver help.
Another Two Views on Disasters
The NoJitter article "Earthquakes, Floods, and Pestilence: Life in the Northeast" by Michael Finneran was written before the Derecho storm and before Sandy. Michael's last comments in his article are still valid: "The biggest lessons from this however should be 1) communications networks truly are critical in these emergencies, 2) people have adopted the new options and will include them in their list of options in difficult times, and 3) the public broadcast media needs to be soundly scolded so that next time, they overcome the urge to sensationalize for ratings purposes and understand their real responsibility is to provide the public with clear, accurate, and useful information."
Another article on the effects of Sandy was included in the IEEE Computing Now publication, "Disaster Recovery in the Wake of Hurricane Sandy: A Personal Account". It is worth a read to get another view of our preparedness or lack thereof for disasters and our ability to sustain adequate communications during and after emergencies.
I think that a public-private initiative, one that is more than talk, will be necessary to create the network services resiliency we already need but do not have.