Study Warns of Internet Brownouts By 2010
Bergkamp10 writes "Consumer and corporate use of the Internet could overload the current capacity and lead to brown-outs in two years unless backbone providers invest billions of dollars in new infrastructure, according to a new study. A flood of new video and other Web content could overwhelm the Net by 2010 unless backbone providers invest up to US $137 billion in new capacity, more than double what service providers plan to invest, according to the study by Nemertes Research Group. In North America alone, backbone investments of $42 billion to $55 billion will be needed in the next three to five years to keep up with demand, Nemertes said. Quoting from the study: 'Our findings indicate that although core fiber and switching/routing resources will scale nicely to support virtually any conceivable user demand, Internet access infrastructure, specifically in North America, will likely cease to be adequate for supporting demand within the next three to five years.' Internet users will create 161 exabytes of new data this year."
I worked for Cisco Systems in the late 90's and through the dot-com bust. Starting in 1995, there was a MASSIVE undertaking to lay out fiber across the nation and throughout the world. When they pulled fiber, they didn't just pull one strand. Fiber is cheap, it is the manual labor that is incredibly expensive to bury the cables and hook them up, certify them, etc. When they buried the cables, they ran 128 pair, 256 pair. TO THIS DAY, we have MORE DARK FIBER than we have lit fiber. There is enough fiber spanning this planet to support a quintupling of bandwidth and we'll STILL have dark fiber to spare.
Why are they 'warning' of impending bandwidth crisis? It's pretty simple.
I was just at a customer site last week (a city government). They had a DS3 and were going to get a second one. I asked him why on earth he was getting a DS3 which is OLD telco technology. I went up to his demarc point and showed him that Qwest had a fiber cable coming into their facility that provided 100mb to the net, that they then fed into a Fujitsu FL4100, then passed it off to a DS3 mux and passed off to the customer as a copper coax connection. They had a wall filled with equipment JUST TO SLOW DOWN THE CONNECTION to a DS3 speed. Oh, and the City was paying for the electricity for all the telco equipment.
I told him to call up Qwest and tell them to come get their crap out of his server room, take the fiber and plug it directly into his switch. And he was only going to pay $2000 a month for the 100mb connection to the internet or else good luck ever getting a permit to dig up another sidewalk in this town.
It worked. He didn't even have to resort to the threats. Qwest knows that they NEED TO CREATE A PROBLEM IN ORDER TO CHARGE FOR THE SOLUTION. In 100% of the cases I've dealt with telco's, I've told them what the speed and feed was that I wanted, and what I was going to pay for it. Never have I had an issue. Now, I do live in the Twin Cities Metro Area, where there is plenty of bandwidth to go around, and I'm not demanding that they give me priority QoS all the way to their tier 1 core backbone, but this game they're playing is ridiculous.
Another customer was paying $12,000 per month to get a 200mb connection to the net. I got on the horn with Qwest and told them to give us a gig connection for $10,000 per month or they can come get their gear because we weren't going to pay for the electricity for them any more. They gave us a gig connection.
It costs $100 to provision a 10mb connection port. Heck fiber optic modules are CHEAP. Want to know how much it costs to reconfigure that link for 100mb? Same Price. It is also the same price to bring it up to a gig connection.
They will bring in equipment for the sake of bringing in equipment, they will spend tens of thousands of dollars in gear just to slow your connection down, just so they can charge to speed it up.
Don't fall for it.
Good security is based upon reality and common sense. Common sense is a function of having common knowledge.