Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×

HD Video Could 'Choke the Internet'? 629

richdun writes "Yahoo! is carrying an AP story explaining how ISPs are worried large streaming videos could 'choke the Internet.' This is used as a yet another reason for tiered pricing for access to content providers." From the article: "Most home Internet use is in brief bursts -- an e-mail here, a Web page there. If people start watching streaming video like they watch TV -- for hours at a time -- that puts a strain on the Internet that it wasn't designed for, ISPs say, and beefing up the Internet's capacity to prevent that will be expensive. To offset that cost, ISPs want to start charging content providers to ensure delivery of large video files, for example."
This discussion has been archived. No new comments can be posted.

HD Video Could 'Choke the Internet'?

Comments Filter:
  • Where I work.. (Score:5, Informative)

    by dadragon ( 177695 ) on Sunday May 14, 2006 @09:24PM (#15331778) Homepage
    Where I work, which is a Canadian telco and ISP, we're doing a major infrastructure upgrade to transmit HD media over our backbone to our DSL subscribers to get IPTV. In October the system is supposed to go live, with 40 meg streams to the house, with a future of 120 meg, and then on to fibre. Quit bitching and develop the infrastructure. It's going to happen sometime anyway.
  • by LightStruk ( 228264 ) on Sunday May 14, 2006 @09:29PM (#15331812)
    What the telcos don't want you to realize is that they are already paid for the use of their wires on a per-packet basis by the owners of the routers that connect to them! Everybody but the consumer pays for the bandwidth they actually use. Today, if an ISP starts sucking down lots of bandwidth because its customers are watching HD TV, the ISP has to shoulder the larger bandwidth bill from the telco. They then pass the costs along to the customers who are using the most bandwidth.

    Google and Joe Webclicker are NOT the telcos' customers! They already pay their ISPs for service. Nobody is getting a free ride.

    The market should drive this process! ISPs that want more bandwidth (so they can deliver hi-def video to their customers) will look for the most bandwidth at the lowest price, and the backbones compete to upgrade their networks so that those ISPs sign up with them.

    Why won't anyone stand up in Congress and say, "but Mr. Verizon, Mr. AT&T, aren't you just trying to charge twice for the same service?"
  • by Jeff DeMaagd ( 2015 ) on Sunday May 14, 2006 @09:36PM (#15331846) Homepage Journal
    Multiply it by 20 MB or so for HD. See how it scales?

    Multibly what by 20MB? Neither the unit nor the number used don't make sense.

    You might have made sense if you had said that HD video can easily consume 4x to 6x the bandwidth of standard definition. And that bandwidth does cost a lot, even with crappy low bandwidth video from YouTube, they don't have a business model to pay for what they are using. They really don't have the media that justifies HD either.
  • Re:Multicast? (Score:4, Informative)

    by Crizp ( 216129 ) <chris@eveley.net> on Sunday May 14, 2006 @09:38PM (#15331860) Homepage
    That's something I've been wondering for years. Not having the knowledge neccessary I still ask: What is the reason anyone and their mother can't set up a multicast audio/video stream? I mean stuff like a 128Kb MP3 stream internet radio station without sucking (128 x N users) Kb in bandwidth?
  • by papasui ( 567265 ) on Sunday May 14, 2006 @09:52PM (#15331913) Homepage
    I'm a Network Engineer for a major US cable company and for about 15 months or so we've been moving our HD streams as IP multicast across our internal fiber network. It's not really that much bandwidth internally to our facilities, about 30 Mbit per channel. Once it reaches our facilities it's converted to QAM and can be streamed across the RF cable plant. Where this could/will pose a problem is for network rider services (ala Vonage) where this traffic needs to cross the egress POP. Anyone involved with carrier level services is well aware that bandwidth is oversold. It has to be due to the insane prices an OC-48 costs. It relies on the assumptions that 1.) Maybe 20-50% of your users will be using the service at any given time. 2.) Even if 100% of your user base is using the service they aren't all using the maximum speed available (ie web browsing versus running Bittorrent). So to sum up, yeah it's not a big deal for a few people to stream HD at 6~10Mbit through an egress point however if a killer service takes off and everybody starts using it in this way it could seriously impact service. In fact it could force a paradigm shift in the industry.
  • by scronline ( 829910 ) on Sunday May 14, 2006 @10:18PM (#15332030) Homepage
    It's only the large, money hungry ISPs doing this. MOST of the ISPs I work with (I'm a board member of CISPA or California ISP Association which ever you prefer) don't like, nor will they practice this kind of crap.

    Personally, the way _I_ see it, I hope they do start doing this. Customers will get angry and find other providers that don't do this. Which means people will go to the better providers anyway.
  • by slappycakes ( 714339 ) on Sunday May 14, 2006 @11:34PM (#15332298)
    I live on a tropical island in the middle of the pacific (Okinawa) and have residential Gigabit fiber for about $70 a month (including land phone). I routinely get 4MB upload and 3MB download simultaneously. There is a business model that works because Japan is using it right now. So, why can't the US figure it out?
  • by SanityInAnarchy ( 655584 ) <ninja@slaphack.com> on Sunday May 14, 2006 @11:43PM (#15332342) Journal
    You're thinking Broadcast, not Multicast.

    Multicast means... Think of a packet as being like an email. It has a To and a From header (sort of). Multicast lets you say "To: 1.2.3.4, 4.3.2.1, 5.1.8.3, ..." You provide a list of ips it's supposed to go to.

    If a lot of those can be reached through the same router, you just send one packet to that router, it splits it up whenever it has to. So, if you have router A connected to routers B and C, you send one packet to router A, it sends one each to routers B and C, and routers B and C split them between whatever clients are requesting it. If no one on router C is requesting it, router A only sends a packet to router B.

    Anyway, look it up, I'm done explaining for now.
  • Re:Multicast? (Score:1, Informative)

    by Anonymous Coward on Sunday May 14, 2006 @11:44PM (#15332344)
    You failed reading comprehension 101 I take it. Multicast is the idea that a client can subscribe to a stream of bytes coming from a provider -- and that each upstream router would be notified of the client's subscription. Assume that both A and B had asked for content from provider C -- and router R was between A, B and C. Content would come out from A to R -- and then R would multicast copies of the packets to A and B. Multiple leaf nodes can talk up to multiple intermediate nodes, on up to the originator.

    The implementation is more complex (dealing with subscribing/unsubscribing, the allocation of the virtual addresses in the 224-239.x.x.x nets, release of the virtual address, timeouts, most protocals also have a backchannel which allow lost packets to be resent by request of the affected client(s) ...)

  • Re:Multicast? (Score:3, Informative)

    by mbone ( 558574 ) on Sunday May 14, 2006 @11:53PM (#15332375)
    True and not true.

    Multicast has developed to the point where there is little doubt that one service model, Single Source Multicast (SSM, explained further at the Multicast FAQ file [multicasttech.com]) could serve unlimited numbers of receivers with a stream, even in the commodity Internet. And Multicast is powering most new IPTV deployments - see the U Wisconsin DATN [wisc.edu] for a cool example. BUT, content providers do not want to supply their content with global SSM multicast, and there is no strong demand yet for sourcing niche video channels. (Existing deployments use multicast to get from a local POP to the user, but do not allow multicasts in from outside.)

    BTW, 3GPP MBMS and 3GPP2 BCMCS now allow for true multicast to wireless phones, but there is as yet little use of it.

    The BBC is trying to change this with their Multicast trials [bbc.co.uk], and I think it almost inevitable that multicast channels will be allowed into the "walled gardens," but end users are only likely to get this ability if there is strong customer demand for it.

    Note, BTW, that multicast in practice won't help an ISP that has severely underprovisioned their edge circuits, at least if there is a typical distribution of channels being watched.
  • by Anonymous Coward on Monday May 15, 2006 @12:06AM (#15332426)
    I agree. Koreans have been enjoying **full-length** feature films, TV shows on-line for several years and ISPs in Korea has had no problem coping with the traffic. Note that all four major networks of Korea (KBS, MBC, SBS and EBS) have been streaming their contents both live and on-demand (for the former, it's free. while for the latter, they charge modest fee). On-demand contents usually come in 100kbps, 300kbps, 700kbps and 1Mbps.
  • by happyemoticon ( 543015 ) on Monday May 15, 2006 @12:12AM (#15332446) Homepage

    Seconded. A member of my family worked on laying fiber for Pac Bell (back when there was such at thing), and the reason they didn't lay nearly as much as they wanted to was just local red tape. Municipalities exert a lot of control over this kind of thing, and not only do they want you to pay to upgrade their city's infrastructure, they want some added perqs too.

    And of course, the same kind of red tape occurs when you want to do anything involving multiple city governments. There's no such thing as, "for the good of the county and region" for these people, there's just their own constituents. And if those constituents happen to be affluent rather than poor or middle class, you're going to have a helluva time getting anything through there.

    Take BART (Bay Area Rapid Transit), for example. I've heard (might be a tall tale, now) that it was supposed to not only go from San Francisco to San Jose, but that it was supposed to go up into Marin County as well. It just didn't happen. They stopped in Millbrae, which is about 12 minutes outside of SF. In order to get San Bruno (the next town in the direction of SF) to allow the rails to go on their land and to the airport, they needed to build them a new police station, and this was only after they were at least four years late.

    And don't get me started about engineers employed by most cities. My closest friend works for the city of San Bruno, and while he was in the water department, the engineers tried to drill a well after the people in the water department said that there was a 90% chance they'd be drilling straight into a sewer main. What did they hit? A sewer main.

  • by hitmark ( 640295 ) on Monday May 15, 2006 @05:11AM (#15333100) Journal
    sorry, i was trying to be funny. should have known it was badly worded...
  • by Anonymous Coward on Monday May 15, 2006 @09:34AM (#15333888)
    They have face to say this crap while there is a technology like Multicast existing for ages.

    Why not implemented on regular ISP? OK no conspiracy theories.
  • by ajs ( 35943 ) <{ajs} {at} {ajs.com}> on Monday May 15, 2006 @09:40AM (#15333930) Homepage Journal
    Time to issue the standard latency vs. bandwith missive, clearly.

    What ISPs are selling is latency. Watch the ads: "the page loads / game plays so fast!"

    They're not selling bandwidth, even though that's how they inaccurately measure their latency. If they were, then servers would not be an issue.

    All of that is moot, however, since there's simple math here:
    rate - usage*cost = margin
    That is, they are selling a service which costs them a certain amount, and they see some percentage usage. They then charge some rate and the delta between those is the profit margin for them. If you are arguing that they should raise the price and eliminate bandwidth concerns, then that's one thing, but if you are suggesting that they keep prices the same, then clearly they have to control one of usage, cost or margin.

    Margins in the ISP business right now aren't spectacular, but they're OK. ISPs certainly aren't looking ot LOWER them, so give up on that point. Then you have usage and cost. The cost is negotiated fairly strongly, but ultimately you have the same argument up-stream with backbones as you have between consumers and their ISPs. Then there's uage. Observe the current trend in attempting to manage usage.

    If you really want to be charged for a full 1.5, 3, 5 or whatever you have down, you're going to have to expect that prices will skyrocket! If that's what you want, then what's wrong with tiered service?

    From where I stand, the whole argument AGAINST tiered service is that the economies of scale in the averaged cost model favor a single tier of consumer service. Then again, I'm a Speakeasy customer now, so I've essentially opted for tiered service anyway by paying more than your average cable Internet user.
  • by Fallingcow ( 213461 ) on Monday May 15, 2006 @05:17PM (#15337929) Homepage
    ALL cable TV on-demand, each subscriber would be getting a single channel (or let's say 3, with DVR dual and tripple tuners all taping something) instead of 350.

    I used to work for a company that did exactly that, with fiber from the main office to hubs and twisted pair copper from the hubs to the users. Ran telephone and DSL over the same line. Pretty slick setup.

    The big down side was that we could only stream 3 channels per line, so someone with, say, 5 TVs and a TIVO would need to pay for two separate lines, or just accept that the 6 devices could only tune to a total of 3 channels at any given time. Of course, satellite and "digital cable" have some of the same limitations.

8 Catfish = 1 Octo-puss

Working...