Network Blackout 183
An anonymous reader writes "Renesys put together a special report on the effects of the recent blackout on routing and network reachability on the Internet. It includes a cool animation of networks dropping off the internet (presumably as a result of the power outage). It is interesting to see how localized some of the outage was--networks in New York state right up to the Vermont border go dark while everything on the other side of the border is quiet. New York City obviously gets clobbered."
Mirror (Score:2, Informative)
For the "Deregulation was the problem" nutjobs (Score:3, Informative)
Some of the finger pointing in the national press has been at deregulation -- if it weren't for deregulation, we would be better able to control and manage the grid. This misguided contention is incorrect in a number of ways.
First, the "deregulation" that has occurred in electricity has primarily been in opening up wholesale markets for power generators and their customers (i.e., utilities), enabling people in Manhattan to continue consuming power (and clamoring now for more regulation) without Con Edison having to build more power plants on the island itself. The existence and growing vitality of wholesale electricity markets has created substantial value in the past decade, through encouraging generation where it is cheapest and sales of power to where it is most needed.
But this limited amount of market liberalization has left the industry in an awkward place. Generation is largely governed by market processes, but transmission and retail distribution remain heavily regulated. The investment decisions of transmission owners and the retail rates that they can charge to their end customers all hinge on rate cases that are decided by state-level regulators. The rates that regulators allow take into account changes in costs, required investments, and the payment to the utility of a rate of return on the assets they own. For much of the past decade this rate of return has been substantially lower than what utilities could earn from doing other things with their money, so they did not invest in building much new transmission capacity or in upgrading existing lines. Nor did a regulatory environment that is a relic from the 1930s, constructed to govern and control local, vertically integrated utilities, either have the incentive or the wherewithal to force the utilities to invest in transmission assets that would carry power to customers in other states.
This lack of investment in the infrastructure that carries the product exchanged in growing, vibrant wholesale electricity markets has become a problem -- not an overnight problem, as those who follow the industry have been concerned about transmission capacity for at least five years. The numbers offered this weekend suggest that electricity volume has increased 30 percent while transmission carrying capacity has increased only 15 percent. This fact illustrates the mismatch between the dynamic markets for wholesale power and the rigid, maladaptive set of state-level regulations and incentives that govern transmission investment decisions.
Markets adapt to changing conditions. The existing electricity regulations do not, and because of that, the transmission infrastructure has not adapted to the increased demand on it from the increasing vibrancy of wholesale electricity markets.
So how do we proceed to ensure that a blackout of this magnitude does not happen again? There are four things that can relieve the strain on the grid. The knee-jerk reaction of many people is "build more wires!" More wires will increase the carrying capacity of the system, and in some cases transmission owners can add lines to existing paths. But this approach faces some serious obstacles -- such construction is expensive and time-consuming. Most importantly, though, getting new lines and towers sited is increasingly difficult, as people and communities object to having such large structures near them or strung overhead.
A second option is to use new technologies, such as high-temperature superconductors and sophisticated computer switching, to upgrade the capacity of the existing power lines. While also expensive, this option gets around the NIMBY issues that accompany the siting of new lines.
A third option is to build
Re:Florida? (Score:3, Informative)
Re:backup? (Score:1, Informative)
Origins of the Internet - no power, no work ? (Score:5, Informative)
Re:Origins of the Internet - no power, no work ? (Score:3, Informative)
Re:Origins of the Internet - no power, no work ? (Score:5, Informative)
-B
Re:Woohoo Toronto (Score:1, Informative)
Re:redundancy (Score:1, Informative)
No. The telcos invested big time to keep the phone system in the 9's. I'm going out on a limb here, but I'll bet anyone who had an old-fashion, line-powered telephone still had a dial tone (and I'm not talking about one of those whiz-bang cordless phones).
MIDS (Score:1, Informative)
weekly [matrix.net] & monthly reacability charts [matrix.net]
and weekly [matrix.net] & monthly [matrix.net] packet loss charts
Their Internet Weather> animations would also be interesting [internetweather.com]
Re:backup? (Score:3, Informative)
I know it's too much to ask here, but I would suggest you read the story.
Re:Slashdotting animation (Score:3, Informative)
Did y'all notice *how* *small* that animation is? Someone else here put it together. I didn't know that animated gifs could compress that much.
Re:Florida? (Score:5, Informative)
There are some interesting, precise tricks that you can play by sending various kinds of packets (usually UDP) and using detailed latency information about each hop of those packets, along with a provider network map, to get closer to the physical location of a particular IP address. We didn't do that for these maps.
Re:redundancy and surge capacity (Score:3, Informative)
We had a presentation last night from Tom O'Rourke [pbs.org] on the critical infrastructure affected by Sept 11 2001.
The main point to come out of it is that most critical infrastructure is commercially run. They are designed to run and handle typical peak loading.
And there is a difference between redundancy and surge capacity. Redundancy allows you to continue operating at a reduced or normal level. Surge capacity allows you to operate at increased levels due to an unexpected event - such as handling increased mobile phone demand by increasing available spectrum bandwidth as was also done after Sept 11 2001.
The problem that you have in these sort of failures is that not only have you lost capacity, but you see increased demand at the same time. One slide he had detailed cell phones in NYC. Typical block rate (no tone) is about 4% - ie one in 25 calls you can't make. After losing all the capacity after the collapse, combined with everyone wanting to talk on their phones, the blocked call rate jumped to 92%. The expense required handle these sort of extreme events cannot be justified.
Utilities can handle ordinary spikes in their systems, but it is not economic to design surge capacity into most systems.
You'll find this problem across a number of sectors... telecommunications, power, and hospital beds tend to provide the best examples. You'd be suprised at how few hospital beds are generally available at any given point in time.
If we had a mass casualty event in New Zealand, it is quite possible we would be sending victims to hospitals in Australia because we don't have the surge capacity.
Cheers Gav
Re:backup? (Score:4, Informative)
Uh, yeah, thanks, we never thought of that.
UPS's run out of juice. Generators run out of fuel. Generators turn out to be less than perfectly maintained and fail after a couple hours. Budgets get trimmed, maintenance gets overlooked, blah blah blah it's never a perfect world. If it was, engineers would be replaced by algorithms.
Companies that are dead serious about power reliability run generator tests every day, and when lightning is detected within miles we automatically start up all the generators and run off them. Yes, we actually go OFF THE GRID every time a thunderstorm rolls in, and in Atlanta that's many times a week. Of course, we've got millions of dollars an hour [turner.com] running through our facility so heads would roll if we weren't this paranoid.
I believe we keep our tanks fueled up for 3 days of continuous service, and we pay a premium to guarantee that when the shit hits the fan, WE get refueled first.
Of course, hardcore multiply-redundant (and *tested*) systems are something that elude the typical IT crew staffed by DeVry grads.
Problems with auto electric generation (Score:3, Informative)
An altenator is different from a generator in that the supplied field is electromagnetic...it depends on another voltage source. This source is supplied (initially) by the battery or residual magnetic field when you start up the car, and is varied by the regulator. When your car needs recharging, the regulator ups the current or voltage in the altenators field wire, and your car cranks out more volts and recharges the battery. When the battery is mostly charged the regulator trims the field down so the alternator produces about 13V.
This has the advantage of not robbing your engine of power and gas milage when the battery has a full charge. However, to use it as a generator, you're going to have to hard wire the field to a constant (and not to the altenator output-- this creates a infinite voltage loop that will kill the altenator).
The reason new car altenators are not ideal is because the voltage regulator is internal to the altenator. This was done for simplicity and to make sure the regulator was always grounded, but it really complicates what we're trying to do. Older cars (until late 70's, early 80's) have external regulators and cars older than that (early to mid 60's ?) had generators (field was a permenant magnet).
Also, you do NOT want to bypass the car battery. It is an excellent power conditioner for the not-so-even generation of the altenator/generator