Once More, With Feeling: The Internet Isn't At Risk Of Running Out Of Bandwidth
from the no-exaflood dept
For years, we’ve been hearing telco execs, telco lobbyists and politicians screaming over the coming death of the internet due to an “exaflood” of bandwidth, as things like internet video and bittorrent totally overwhelmed the internet infrastructure. There was little proof that this was actually an issue, and plenty of evidence suggesting that ordinary infrastructure upgrades would more than handle all expected growth. And, in the last few months we’ve been seeing more and more public reports supporting this position. In August alone we saw two separate reports noting that internet growth was actually slowing rather than increasing at an alarming rate.
And now there’s a third such report, looking at internet backbone traffic and noting that there’s little to worry about:
For the second consecutive year, the rate of underlying international Internet capacity deployment outpaced global Internet traffic growth, leading to lower utilization levels on many Internet backbones. Between 2007 and 2008, average traffic utilization levels decreased from 31 percent to 29 percent while peak utilization fell from 44 percent to 43 percent.
Yet, if you listen to telco lobbyists, execs and politicians, they’d have you believe that over the past couple of years, the growth of BitTorrent and internet video was flooding the networks. Hopefully, with so many reports pointing out the opposite, politicians will finally start pushing back the next time a lobbyist or exec starts claiming that the internet is at risk of running out of bandwidth.
Filed Under: bandwidth, bandwidth crunch, exaflood, studies
Comments on “Once More, With Feeling: The Internet Isn't At Risk Of Running Out Of Bandwidth”
It all comes down to companies being cheap assed bastards. Comcast has billions and cant spend a few million to upgrade their infrastructure so they dont have to cap, throttle, and traffic shape traffic? How about they stop advertising to new customers that they cant support and use that money to upgrade their infrastructure
Re: Re:
I’ve mentioned it before. It’s not a matter of whether they want to spend the money to upgrade their infrastructure (which they will anyways, because eventually better technology will outweigh the cost of implementing it, because it will save them money in the long run) but more a matter of applying a new business model that they can control.
They’ll begin selling tiered cap models as a “premium” for being with them. The end result: more customers come in due to lower cost internet service plans that are capped at very small levels, while “premium” caps will sell for much more than what can currently be had on the “unlimited” service model.
It’s a dickheaded thing to do, but it’s really no different from telco’s feeding customer’s crap arguments like “texting uses excessive bandwidth and overloads our infrastructure, thus we need to charge extra for the service”. Bull.
Politicians not parroting lobbyists?
Pah, I say. The odds of a politician not being frightened and not amused at the phrase “exaflood” are exceedingly low. In technological issues, at least, they have little to no credibility, and less knowledge of how to legislate on these issues.
It’s sad and funny at the same time.
This is sick – but I see a possible problem. Can I ask what the resolution a human can see at is? I know the human eye has about 100 million receptors. With current HD, video is 720 by 1080, right? Now, that’s 3/4 of a million pixels. Assuming that the eye uses only 1/4 of those receptors, that would require a resolution of 5k by 5k, or I suppose something like 4k by 6k, really. Now, assuming compression rates stay the same with bigger video, that gives us about 31 times the current size for video.
Assuming OLED’s are miraculous, tiny, and the screen is the only thing in your vision, of course.
Hopefully we can coax more bandwidth out of fiber optics.
I’m just speculating.
Re: Re:
I am at a loss for worlds on this Epic Failure of a post
Politicains will do what the money tells them to do.
I doubt that they are free to make logical decisions.
You’re missing the point. This is the beginning of Telco and Cable to start setting usage limits and raising prices. Comcast will in very short order start pricing their bandwidth like wireless companies price cell-phone usage.
X Gigabytes of bandwidth, 39.95/mo. Family plans with more bandwidth, 99.95/mo. or whatever. Want movies, add a few more dollars a month.
these idea is an idea only, i think a good reason why bandwidth is running out is cause by spammers and viruses. sending unwanted files. like a garbage blocking a sewer
—
Josephine Swann Email Address Directory
did u read?
It says that we are not running out of bandwidth white pages. Also I think its stupid that comcast and others are limiting bandwidth. How bout give real internet speeds and then when you realize that people like to download things and we truly are low on bandwidth then you can cap it. Till then I think any limit on what was cheaper unlimited is a very bad thing and that should cause riots. Y is no one freaking out about this???
Re: did u read?
I’m freaking out about this. This is tactic is like boiling a live frog. Incremental changes go unnoticed until it’s too late.
about:internets
surly they aren’t wrong?
What is your definition of “internet growth”? FGrowth of the infrastructure or growth of the total amount of traffic? Theres a big difference.
It sounds to me like both are rising. Infrastructure is rising faster than traffic, so the utilization is slowing.
An investment driven straw man.
Once again, with feeling 😉
A report on the core of the network tells you about the core of the network. You can not reach any conclusions on other parts/layers of the network. There is insufficient data to do so.
Here is a straw man arguing that it is feasible that there is a problem, just not in the core.
Enormous amounts of money were invested in core fibre networks and submarine networks the late 90’s and early 00’s. There was so much oversupply in core infrastructure that demand is really only just starting to catch up. Even after all of the chapter 11 restructuring, outright bankruptcies and industry consolidation earlier this decade the networks have continued to invest in core infrastructure.
I used to work for a (non US) Telco building that core infrastructure.
So another report measuring the utilisation of the core of the internet is concluding that there is no congestion in the core. Hardly a revelation.
The question though, is to what level did the combatants invest in the last mile, the local access and regional networks connecting the consumer to the core of the network? Did it come anywhere close to the billions that Enron, etc. frittered away?
I suspect not.
The conclusion: comcast et.al. are facing problems (or potential problems) in their broadband access networks through a lack of investment there. The same problems are not evident in the core because of early over investment and sufficient on going investments.
OK. Knock me down.
Re: An investment driven straw man.
A report on the core of the network tells you about the core of the network. You can not reach any conclusions on other parts/layers of the network. There is insufficient data to do so.
But, the complaints about an “exaflood” aren’t about the end points of the network, they’re about the core.
And, as plenty of others have shown, if the core network is there, you can upgrade that last mile to be able to handle much more traffic.
That’s the point. Those basic upgrades will do more than enough to handle the traffic.
Re: Re: An investment driven straw man.
But the complaints about an ‘exaflood’ aren’t about the end points of the network
Maybe I’ve read different papers:
I’ve been basing my responses to your posts on the above.
Perhaps
the suits when talking about bandwidth shortage, they actually mean the last mile. 🙂
Nah, unlikely, that’d be customer driven, and it’s quite clear that ISP’s couldn’t care less about their customers.
An investment driven straw man.
Allen wrote:
“The conclusion: comcast et.al. are facing problems (or potential problems) in their broadband access networks through a lack of investment there. The same problems are not evident in the core because of early over investment and sufficient on going investments.”
You get no argument from me. I would have said this has been noted here many times before. It is a last mile problem. Look at it from the providers perspective. They have a problem, there is not enough bandwidth in the last mile (read their equipment/cable) to provide each user with all the bandwidth they want. The users see a bottleneck. From most users/congress critters understanding the internet is one homogeneous thing. So the providers have a choice, they can spend money to improve their system, and make less profit, or they can use the average individuals assumptions against them, claim insufficient bandwidth (not really saying where), cap users (“what else can we do, the internet is full”), and cry at the government for relief.
They are after all a for profit company, not a public utility. They have to make a profit, at the consumers expense. Caps are more profitable than providing the hardware to give the consumer what they were told the buying. Statistics are a wonderful thing. “Your bandwidth is unlimited, as long as you only ask for the statistically average amount of bandwidth we expect you use.”
Can you guys speak up?
I cant hear you, my interweb pipes are all clogged up with bittorrents . . .
Politician push back lobbyists?
I want two of whatever you’re drinking ;0)
Looming crisis
The declining utilization rates of the internet infrastructure is a looming crisis for our industry. To ensure that the infrastructure continues to be supported and meets the continuing needs of Americans everywhere, the industry is forced to ask for an increase in our tariffs and for further government grants to support this critical national resource.
Thank-you for your understanding.
Sound like
This sound very similar to global warming. Just politicians looking for something else to tax.