It is quickly becoming widely understood that voice as VoIP is just another application like e-mail. Its a no-brainer, yesterdays news, right? Not such a no-brainer a few years ago though. So, what makes video any different? If the differences in the protocols and compression, such as G.711 versus MPEG4 are set aside, nothing is different but the bandwidth requirements. Video is just another application as well. IP is peered, VoIP is peered, and video is next.
Video peering is the Wild West and, just as ENUM within VoIP (define - news -alerts) Peering is disrupting the economics of the traditional voice business model, the mere capability of a globally distributed IP network openly carrying video feeds, clips, and feature-length films poses real threats to broadcast television as well as the cable companies. Its no wonder that the RBOCs are racing with lightning speed into IPTV. Dont be surprised when they all begin to peer with each other to deliver video content not that theyre moving anywhere near as fast with VoIP Peering and ENUM. Knowing the players, their vantage point, the revenue models they must protect, and the ones they must attack, sheds a whole lot of light on their motivation.
VoIP as just another application trivializes and compartmentalizes an entire industry that for, 100+ years, employed and fed hundreds of thousands of people. Such is life, such is evolution. VoIP is becoming a component of every type of IP interaction imaginable, starting with online reservations and heading straight in to the audio component of a full-duplex video session. Just when you thought the rate per minute couldnt go any lower, in came flat rate voice. It doesnt end there. Now voice as audio is just another component of video like color versus black and white. When you watch TV, you dont have to pay extra for sound.
With the exception of a few sparsely inhabited places on Earth and those with governments that inhibit development and new economic growth by nationalizing ENUM under a legacy revenue model, there will be no separate business model for the audio piece. It is embedded and a given. After all, most of what will be driving the voice piece in the IP world will be free, open source software. Since this is now becoming clearer, we must revisit the bandwidth requirements for video.
Video over IP does not mean video over the Internet in exactly the same way that VoIP doesnt mean voice over the Internet. This is very important and it frames the critical debate of IP Networks versus The Internet as the transport medium. With voice as the main character the debate for the carriers is split. Some want to further utilize their existing ISP investments, while others see the benefits of security and reliability outweighing any incremental costs associated with operating or using a private voice Internet. Many small to medium enterprises have no concept of private VoIP connections to other businesses and still see voice as a service delivered to them by a provider rather than the application that it is, such as e-mail, riding over an extended WAN. For large enterprises there is no VoIP over the Internet debate. The answer is no. Private wide-area network connections carrying the VoIP traffic is the only way to go. Cost is not the concern, whereas security, performance and reliability are.
These are all very prudent concerns and each ultimately leads to its own respective, logical conclusions.
Whats interesting about the public versus private VoIP debate is that from a capacity supply versus demand perspective it is actually possible to have it in the first place. This is because the most bandwidth conservative CODEC out there, G.711, consumes only a mere 64k for a full-duplex conversation. That means that the Internet can, and does, successfully carry high-quality voice calls. This is due to the underlying ISPs all having OC-192, or close, backbones. Millions of simultaneous G.711s dont amount to much strain within 10 Gig. If we look at the more aggressive CODECS such as G.729 the compression gets better and the number of calls can increase without impacting quality too much.
With video as the main character, the story changes considerably. Those old 192 backbones are no match for real-time video when it comes to servicing a multiplicity of networks and users simultaneously. One network alone could consume a Gig. That tops out the 192 at 10 networks. Simply put, the public Internet is not ready, nor are the ISPs in a position, to accept the several orders of magnitude increase in bandwidth required to support the IP enablement of live broadcast video. This is not to say that canned video in conjunction with a TIVO cant be supported it can, but even those files sizes are already increasingly putting a strain on whats out there today.
This pending bottleneck limitation is not slowing the pace of development, though. So, how is the issue addressed? Well, peering, of course. Not peering via the public Internet, but rather peering as the ISPs do to CREATE the Internet itself. Video networks (content and plain transport) using dark fiber, waves and Ethernet will all find common physical layer points on the Earth to connect to a common Ethernet switch, establish VLANs between and amongst each other and peer. This is what the ISPs have been doing for a long time, what the VoIP networks are doing now, and this is what the video networks will be doing. Its all a matter of time. IT
Hunter Newby is chief strategy officer at telx. For more information, please visit www.telx.com (news - alerts).
If you are interested in purchasing reprints of this article (in either print or PDF format), please visit Reprint Management Services online at www.reprintbuyer.comor contact a representative via e-mail at firstname.lastname@example.org or by phone at 800-290-5460.