Backbone Analysis Puts Exaflood Myth To BedWireless growth exploding, but wireline growth may actually be slowing... 01:07PM Tuesday Nov 25 2008 by Karl Bodetags: business · bandwidth · Op/Ed · telco · stats · world · networkingLast week we explored how companies like AT&T use cherry picked data, think tanks and policy groups like the Internet Innovation Alliance to promote the idea that we're facing a horrible bandwidth crisis (aka the "exaflood"). According to these lobbyists and PR wizards, this bandwidth apocalypse can only be avoided if you give incumbent ISPs what they want; namely lower taxes, government subsidies, less regulation (or more regulation if it helps them), and the right to implement metered billing. Be it telecom or any other field, bad ideas are usually more easily pushed forth under a climate of fear. But once again, data from outside of the world of incumbent ISP spin suggests that any future growth can easily be handled by reasonable traffic upgrades. Over the weekend the University of Minnesota's Minnesota Internet Traffic Studies (MINTS) compiled the latest set of data concerning the growth of Internet traffic from a flurry of global sources across a slew of different backbones. According to MINTS, while wireless data growth is exploding, wireline network traffic is increasing at a fairly modest clip: 50 to 60% per year. That's substantially less than last week's report by AT&T, which predicted that sustained growth of 100% or more would result in Internet "brownouts" over the next few years. "The basic, and highly debatable, assumption behind (the Nemertes data), though, is that traffic is growing at 100% per year or more, and will continue to do so for the next half a dozen years," says the report. "So far there is little evidence of that, though." In fact, the evidence indicates if anything, global traffic may be declining from the 50-60% annual growth rate. European Internet exchanges saw a decline in annual growth rate of 56% over the past year -- down from 84% the year before. TERENA, a consortium of European national research and education networks, showed a growth rate (pdf) of just 46% per year, with signs of a slowdown. Equinix reported that in the U.S., its total network access traffic grew 34% compared to the previous year, while Cogent -- who last quarter saw their first ever traffic decline, saw traffic grow 5% compared to the second quarter, and 24% compared to a year earlier. Again, there is absolutely no evidence that Internet growth is accelerating so quickly that carriers can't manage it with modest capacity improvements. Related:- Monday Morning Links
- Tuesday Evening Links
- Thursday Morning Links
- Thursday Evening Links
- Friday Evening Links
- Wednesday Evening Links
- Friday Evening Links
- Verizon's Open Development Initiative? So Far It's A Joke
|
  swhitney2003 I can't drive 55. Premium join:2003-06-13 NH clubs:  | just wait Just you wait until there isn't much more internets to go around. Eventually we will have to go out to Californie-way to get some. There will only be one computer, and we'd all have to share. | |
|  |  |  |  |   Matt You can't fix stupid Premium join:2003-07-20 Jamestown, NC
·North State Commun..
·Corporate Colocation
| Re: Past statistics don't always show future trends said by TK Junk Mail :compiled the latest set of data concerning the growth of Internet traffic Those predicting drastic growth may be more prescient than the statisticians looking backward. So the answer is obviously to institute caps? I meant, if bandwidth use is growing as Comcast, Time Warner, and AT&T insist, then clearly capping monthly bandwidth and charging for people who go over that amount is the answer. | |
|  |  |  ggultra2764
join:2007-09-13 Cambridge, NY
·Verizon Wireless B..
| Re: Past statistics don't always show future trends Yeah, but the trust of the three above is questionable with AT&T bugging the government for teleco immunity from lawsuits, Comcast butting heads with the FCC and angry consumers over P2P throttling and invisible caps, and some of Time Warner Cable's decisions regarding channel packages. | |
|  |  |   jmn1207 Premium join:2000-07-19 Reston, VA
·Verizon FIOS
| Caps and ridiculously priced overage charges speak volumes toward most ISP's future business model.
They found a way to increase the transfer speeds in a relatively inexpensive manner, but rather than spend the money on the infrastructure to support these increases, they would rather place a moratorium on the bandwidth consumption limits. They want to impose a freeze on the amount of data that can be transferred. It costs too much money in their eyes. And if they all join forces, this will be the accepted standard and it will not come back to haunt them.
They are trying to make it impossible for innovation to take root. They are crippling one type of service to keep another type alive. This is par for the course with these giant entities that have way too much control and influence in any industry. I wish it were easier to identify and eliminate these nefarious business practices that the RIAA and their ilk seem to employ after they gain too much control. | |
|  |  |   not me 2
@comcast.net
| Whether most people would choose to watch TV online is hard to say. It hasn't happened so far and even if everything was available in HD I don't know that would make many more people switch to watching online instead of on their TV. Online viewing is nice for short clips like Youtube carries, but most people will want to watch on their larger TV screen than their laptop or desktop computer has. (Yes, you can connect a PC to the TV but most people don't do that and aren't likely to do it in the near future.) People are creatures of habit and watching shows/movies on the TV is what they are used to and most aren't going to change that viewing habit, especially not if they have more than one viewing at a time.
In either case it probably doesn't matter because monthly usage caps are here and they aren't likely to go away. The caps will keep people from switching to watching large amounts of HD video online even if the cap would allow someone to do so. That's just because most people don't want to risk going over the cap so they will turn to other sources (watch live or recorded TV) instead of watching online. | |
|  |  |  |  Lazlow
join:2006-08-07 Saint Louis, MO
| Re: Past statistics don't always show future trends not me 2
I think you are VASTLY underestimating the general public. PVRs are very popular, mythtv(among others) is a fairly common thing to see in people under 30 households. Since the computer is already hooked to the TV it usually does not take very long for these people to figure out that they can get media online. It may start out as just finding episodes that were missed for one reason or another, but it soon spreads. If you like one show from a pay channel (say Dexter). If you like a series from overseas (say Kingdom). After a while people soon start to ask themselves why they are paying $60+ for cable when they can get the same media online. While this has been climbing slowly over the last decade(I had my first pvr in 93) it has hit a huge growth spurt since DVRs have become popular (people do not like paying that monthly fee, just like when they got rid of cable boxes the first time). | |
|  |  |  |  |   TK Junk Mail Go ahead, make my day Premium join:2002-03-03 Margate City, NJ clubs:
·Comcast
| Re: Past statistics don't always show future trends said by fireflier :Assuming tremendous growth of high def video really isn't any more accurate than using past years' data to predict future growth. Who can say with certainty what HD video is actually going to do on the internet. How about story after story about Apple TV; Roku; Sling; Xbox; PS3 and now the Blockbuster STB being discussed in the BBR story right after this one.
It is anecdotal evidence right now, but it does speak to a change coming and coming very soon. A massive change in user actions in TV watching is being predicted here and everywhere on the net. And the statistics doesn't show it yet, but predictions say it is coming. -- My BLOG .. .. Internet News .. .. My Web Page Ask yourself one question: 'Do I feel lucky?' Well, do ya punk? | |
|  |  |  |   jmn1207 Premium join:2000-07-19 Reston, VA | Re: Past statistics don't always show future trends Is this anything like the massive change in user actions with music listening? Are we going to have to suffer through a bunch of control freaks that will stifle innovation and hold firm to an outdated business model until the very end? | |
|  |  |  |  |  |  |  |  |   jmn1207 Premium join:2000-07-19 Reston, VA
·Verizon FIOS
| Re: Past statistics don't always show future trends said by fireflier :They've survived other internet advancements (graphic heavy pages, flash, shockwave, MP3, etc) without these conditions. As an apparent advocate of free market capitalism, you should be suggesting the telcos deal with what's coming and let the strongest and smartest survive. Unfortunately the strongest and biggest have decided they want to place crippling caps on our service. And in Canada, the biggest and most irritating is allowed to make sure that all competition must follow their lead in throttling.
There is no legitimate free market capitalism here. Not in the sense where it is beneficial or advantageous to the consumers. | |
|  |   Nightshade sic semper tyrannis Premium join:2002-05-26 Salem, OR
| said by TK Junk Mail :Those predicting drastic growth may be more prescient than the statisticians looking backward. I have a very serious problem with that statement and here's why.
I took statistics and probability. If there's anything I learned out of that class is that statisticians look at past data to make predictions about future trends. Without past data, there is no baseline for them to show future trends.
You can't have one without the other. You got to have past data to create any statistical trends at all regardless if the trend is accurate or not, period. -- Be careful who you vote for, you just might get it. | |
|  |  |   TK Junk Mail Go ahead, make my day Premium join:2002-03-03 Margate City, NJ clubs:
·Comcast
| Re: Past statistics don't always show future trends said by Nightshade :I took statistics and probability. If there's anything I learned out of that class is that statisticians look at past data to make predictions about future trends. Without past data, there is no baseline for them to show future trends. You can't have one without the other. You got to have past data to create any statistical trends at all regardless if the trend is accurate or not, period. So did I. And there is no statistical models that project when the knee in a curve may happen. And that is what you can't predict based purely on past trends. -- My BLOG .. .. Internet News .. .. My Web Page Ask yourself one question: 'Do I feel lucky?' Well, do ya punk? | |
|  |  |  |   Nightshade sic semper tyrannis Premium join:2002-05-26 Salem, OR
edit: November 25th, @03:00PM
| Re: Past statistics don't always show future trends Oh I know. If anything creating statistics is really a combination of mathematical theory and good old fashioned educated guessing. All statistical models will eventually fail, it's just a matter of when and how the data that already exists is modeled. But you still need strong past data to make the predictions. Without that all you really are doing is doing baseless guessing. -- Be careful who you vote for, you just might get it. | |
|  |   funchords Robb Topolski Premium,MVM join:2001-03-11 Hillsboro, OR
·Verizon Online DSL
·Skype
·Comcast
| said by TK Junk Mail :compiled the latest set of data concerning the growth of Internet traffic A lot of these predictions that the internet WON'T face much higher bandwidth demand is based on past statistics. But a paradigm break is in process. The past statistics can't recognize the tremendous growth of online high def video that is coming. Those predicting drastic growth may be more prescient than the statisticians looking backward. I think we're in the middle of that, now.
Again, folks, know your history or it will come to know you (in a Biblical way). 
When the web took off, everything strained under the addition of images,
and then animated images,
and of course demand of the web increased and another cycle was born --
flash and light video and larger images --
multiple computers per household -- more consumers -- more worldwide adoption.
Plus, don't forget that video itself has undergone a revolution. There aren't many 19-inch 4x3 format TV's selling these days.
That's not the future -- that's now. Those that are throttling are throttling the inevitable --
-- they're throttling progress --
-- and more insightful countries are passing us.
How can we invent the technology of tomorrow when throttling ISPs keep us all using the technology of yesterday?
Holy crap! -- Robb Topolski -= funchords.com =- Hillsboro, Oregon More features, more fun, Join BroadbandReports.com, it's free...
| |
|  |  |   jmn1207 Premium join:2000-07-19 Reston, VA
·Verizon FIOS
| Re: Past statistics don't always show future trends It's simple. Some CEO can make $18 million now instead of $4 million. If the company crashes and burns while crippling an entire industry, certainly the people will bail them out with our taxes. Although the tax payer will still be left with an inadequate, overpriced infrastructure with no other options available. | |
|  |  |  |  |   TK Junk Mail Go ahead, make my day Premium join:2002-03-03 Margate City, NJ clubs:
·Comcast
edit: November 25th, @09:27PM
| Re: Past statistics don't always show future trends said by fatness :I realize you've said that before, but it reads more like a provider's press release than empirical information. If Karl can keep posting the same story over and over, I figure I can post the same reply as well. -- My BLOG .. .. Internet News .. .. My Web Page Ask yourself one question: 'Do I feel lucky?' Well, do ya punk? | |
|  |   NetAdmin
join:2008-05-22
| said by TK Junk Mail :A lot of these predictions that the internet WON'T face much higher bandwidth demand is based on past statistics. Using proper forecast models, you can use past data to make fairly accurate forecasts. That's how planning is done on bandwidth planning. It is also how the weather is forecast.
It is true that past performance is not indicative of future performance, but without having a handle on past performance, you can't make an educated estimate of future performance. -- --- Drilling for more oil is akin to giving a methhead the keys to the meth lab. | |
|  |   Brett Schulte
@sbcglobal.net
| said by TK Junk Mail :compiled the latest set of data concerning the growth of Internet traffic A lot of these predictions that the internet WON'T face much higher bandwidth demand is based on past statistics. But a paradigm break is in process. The past statistics can't recognize the tremendous growth of online high def video that is coming. Those predicting drastic growth may be more prescient than the statisticians looking backward. I'm with you - everyone seems to be adding broadband "On Demand" to their Direct TV, Netflix to their Xbox, etc. I see a crunch coming as soon as this holiday... it's not hard to imagine bandwidth would at least triple using these new services... people who go from minimal email and web use are now downloading multi-gig movies! | |
|  kontos xyzzy
join:2001-10-04 West Henrietta, NY
| Woo Hoo! Free capacity! said by Karl Bode :Again, there is absolutely no evidence that Internet growth is accelerating so quickly that carriers can't manage it with modest capacity improvements. Ah, so as long as the capacity improvements are "modest" The carriers don't have to pay for them. Got it. | |
|  |  sjr
join:2006-08-27 Osseo, MN
| Re: Woo Hoo! Free capacity! Interestingly that is how most businesses run. They use part of their profits to maintain and expand their infrastructure. It is what one would call investing in the future. But I suppose if one can only see the next quarters profits one would never expand since that costs immediate money, never mind the payoff down the road in a year or 2 or more. Since the telecom's and cable companies seem to operate in a world of handouts and increasing investor profits every quarter they are not very good at planning ahead. | |
|  |   Dogfather Altitude is your friend Premium join:2007-12-26 Laguna Hills, CA | They never pay for them, their customers do. | |
|  |  |  kontos xyzzy
join:2001-10-04 West Henrietta, NY
edit: November 25th, @03:46PM
| Re: Woo Hoo! Free capacity! I'd agree with you in a metered billing price model. With that model, increasing capacity increases revenue (as long as you can get your customers to use the increased capacity).
But that's not where we're at today. Today we have the all you can eat price model.
All you can eat pricing works great when you're adding customers (revenue) faster that you're adding capacity (costs). The ISP bean counters are starting to crap their pants now because customer adds are slowing down while costs continue to rise. Their graphs are showing the cost and revenue lines starting to converge.
All that the study says is that the convergence may not happen as soon. | |
|  |  |  |  |  |  |  |  |  |  |  |  |  |  |  Lazlow
join:2006-08-07 Saint Louis, MO
| Re: Woo Hoo! Free capacity! espaeth
The part you are forgetting (and ISPs are trying to hide) is that the cost per/unit has been and will continue to drop dramatically. We can all see from the stockholders reports that ISPs are making huge profits. Will expanding the capacity of their networks cost a lot of money? Certainly. But if they quit looking at next quarters numbers and instead focus on the numbers over the next few years there will be no issue. | |
|  |  |  |  |  |  |  |  |  |  |  |  |  |  |  Lazlow
join:2006-08-07 Saint Louis, MO
| Re: Woo Hoo! Free capacity! "You want to upgrade to DOCSIS 3.0, you need to swap out every single cable modem on the plant to get the true full benefit."
espaeth
That is patently false. Docsis 3.0 is backward compatible to previous versions. The only cable modems that need to be replaced are those of the high speed users. You bond their channels off the channels(on different channels than) used by the earlier Docsis versions. This gives the high speed users (presumably high capacity as well) a clear path and it drops them off the capacity load on the channels being used by the older Docsis versions (more bandwidth for the Docsis 2.0 users to share). Yes, only those users that have the Docsis 3.0 modems will be able to get the speeds above 30meg(I think that is the limit for 2.0). But most of the ISPs are claiming that only a small percentage of their users are soaking up the majority of the bandwidth. Presumably these(high capacity) user are usually the high speed users. Even if all the above was not true, every single piece of equipment you mentioned is cheaper today than it was a year ago and a year from now it will be even cheaper. So as time goes on the price/unit drops regardless. | |
|  |  |  |  |  |  |  |  |  See 10 replies to this post | |
 |  |  |  |  |  |  |  |  |  |  |  |  See 12 replies to this post | |
 nath12
join:2008-10-13 Deepwater, MO | revenues Telcos are just trying to quell free high def video so the can keep the triple play going forward. It's just another ploy to keep revenues up. | |
|   IHM Premium,VIP join:2001-12-18 Hamilton, ON
| The last mile... Karl forgets that the last mile is the choke point and keeps focusing on undersold backbones. Upgrading the capacity of the last mile is millions of dollars with little to no return.
I don't understand what is so wrong with paying for the bandwidth you use? The "all you can eat buffet" is a terrible business plan unless you're serving garbage to senior citizens. I'd suggest that metered bandwidth would lower prices for most people AND improve network performance. -- 2 Large Bunnies...1 Dead Penguin | |
|  |  SilverSurfer
join:2007-08-19
moderated: November 25th, @03:21PM
| Re: The last mile... said by IHM : I'd suggest that metered bandwidth would lower prices for most people AND improve network performance. Metered bandwidth isn't lowering the price of anyone's monthly bill except in the myths peddled by providers & their lobbyists. | |
|  |  Qixotl
join:2002-02-08 New Milford, CT
edit: November 25th, @02:27PM
| said by IHM :I'd suggest that metered bandwidth would lower prices for most people AND improve network performance. Has anyone actually seen evidence of price lowering by companies that have already started metering bandwidth? Everything I have read about Comcast, Bell Sympatico, TW, and AT&T suggests that the pre-cap prices are staying the same for their packages once the caps are in place. The addition of overage fees can be considered a price increase to some extent. I just don't trust the providers to pass any savings to the customer unless they are forced to do it by real competition or government mandate. | |
|  |  |   IHM Premium,VIP join:2001-12-18 Hamilton, ON
| Re: The last mile... said by Qixotl :said by IHM :I'd suggest that metered bandwidth would lower prices for most people AND improve network performance. Has anyone actually seen evidence of price lowering by companies that have already started metering bandwidth? Everything I have read about Comcast, Bell Sympatico, TW, and AT&T suggests that the pre-cap prices are staying the same for their packages once the caps are in place. The addition of overage fees can be considered a price increase to some extent. I just don't trust the providers to pass any savings to the customer unless they are forced to do it by real competition or government mandate. That's because the current caps are far in excess of what most users consume. If ISP's sold connections and bandwidth separately, my suggestion is that it would cost less for most users. i.e. truly metered bandwidth at $x amount for the connection, bandwidth billed at $y per GB. -- 2 Large Bunnies...1 Dead Penguin | |
|  |   jmn1207 Premium join:2000-07-19 Reston, VA
·Verizon FIOS
| If it were possible to be charged $0-$100,000 I would love paying for the bandwidth I use. Unfortunately, the stand that most ISP's have taken is to have a high initial set price and then add even higher overage charges on top of this already expensive flat rate.
We will never have a pay as you go model with exorbitant starting costs in place before any usage amount is ever factored into the equation. It's an abusive pricing model and hardly anyone has a realistic alternative. | |
|  |  |  Lazlow
join:2006-08-07 Saint Louis, MO
| Re: Methodology problem And why is it that they only use public data? Because the ISPs will not let their data out. Just look how long it took to get bell to release their data. Look the the BS story comcast came up with for the FCC request (oh it would take too many hours, funny how everybody else got it done). One of the rules the new FCC should initiate is transparency. Force the ISPs to show what their data load is. There is absolutely no security risk in doing this. The only risk is that people will be able to see how much BS has been spread and how much greed is actually going on. | |
|  |  |  |  |  |  |  Lazlow
join:2006-08-07 Saint Louis, MO
edit: November 25th, @11:14PM
| Re: Methodology problem Its funny, in one line you are saying that it cannot be done(safely) and then a couple of lines down you admit that the Japanese are already doing it. To publish this data you do not need to provide attack addresses, I guess I thought that was obvious.
As far as the public data does; there is no real way to tell how valid it is UNTIL somebody (trustworthy) gets the ISP's data. | |
|  |  |  |  |  |  |  |  |  |  |  Lazlow
join:2006-08-07 Saint Louis, MO
| Re: Methodology problem I agree that it should be industry wide. But each ISP will have to supply that data. There is just no way around that. I am not convinced of the necessity of reporting it as a group. I can see labeling is as ISP A, ISP B, etc. But if you pour all the data in a pot and mix it up, you are not going to be able to draw nearly as much information from the data. | |
|
|
|