Why is this page text-only?

ABOUT

App-Rising.com covers the development and adoption of broadband applications, the deployment of and need for broadband networks, and the demands placed on policy to adapt to the revolutionary opportunities made possible by the Internet.

App-Rising.com is written by Geoff Daily, a DC-based technology journalist, broadband activist, marketing consultant, and Internet entrepreneur.

App-Rising.com is supported in part by AT&T;, however all views and opinions expressed herein are solely my own.

« Streaming Media Industry Laments Lack of Broadband, Shrugs Its Shoulders | Main | VidChat: Talking Videocalls with Peter Csathy, CEO of SightSpeed »

May 23, 2008 9:34 AM

The Potential of a Multicast-Enabled Future

During my travels to NYC this week I had the great fortune to meet and commune with Dom Robinson, a fellow StreamingMedia.com contributor and head of GlobalMix, a London-based content delivery network focused primarily on enabling the delivery of live webcasts.

While his interests range far afield, his longtime passion has been for multicast.

First it's important to understand that we live in a primarily unicast world. If you want to stream a live event you first send the video from the location of the event to a server, and then for as each new user shows up you need to send out a new stream. Therefore with each additional user you need more server and bandwidth capacity. So if 100 people request a 1Mbps stream from the same server, that means you need the server to be powerful enough to handle delivering that much simultaneous traffic and you need more than 100Mbps upstream connectivity to serve all those people. (I'm simplifying this dramatically as very few live streams run through a single server that everyone's logging on to, but it gets the point across.)

That's unicast. In a multicast world, things are remarkably different.

Multicasting is akin to broadcasting where you're sending out a single stream that an infinitely large audience can view. When a TV broadcaster sends video over the airwaves or a cable operator sends channels over copper, there's no additional cost or need for additional capacity to support additional users. You're able to just put out a single multicast stream and an infinitely large audience can view it.

The multicast concept is far from new, and I've long known about its similarities with broadcast TV, but in talking to Dom, he opened my eyes to what multicast can really mean. His vision is that some day you'll be able to turn on a camera from anywhere and deliver live video to an infinitely large audience without it costing much of anything to send.

Currently to reach a large audience you need to either setup your own servers or, more typically, employ the services of a content delivery network, which host banks of servers around the world and promise to make scaling up to larger audiences painless. But there's still significant costs involved, which only go up as the size of the audience increases.

Multicasting could enable a world where you don't need to do any of that to deliver live video to a worldwide and expandable audience.

Multicast has not made any significant headway, though, despite being around for years as it's not as simple as just turning on a camera. To work properly, multicast requires the involvement of pretty much everyone in the value chain of content delivery, in particular network operators. They need to enable their networks for multicasting.

I learned in talking with Dom that most networks already do have the capability to enable multicasting, it just isn't turned on. But unfortunately it won't be as easy to get going as flipping a switch as to really work you need to get buy in from every network operator as otherwise a multicast stream will stop working the moment it jumps from a multicast-enabled network to one that isn't.

Many in the streaming industry have given up on the possibilities of multicast because of the general lack of enthusiasm among network operators to implement it and the great challenges multicast faces in trying to get everyone on board.

But I can't ignore how exciting a future could be where I can start streaming live to an infinite audience at little to no cost.

To that end, I began discussing with Dom the possibilities of trying out some test deployments among a limited number of network operators. He couldn't have been more excited, so I'd like to put out an invitation to any network operators out there: if you want to show the value of your network, prove your commitment to enabling all that is possible online, and want to have some fun trying things out on the cutting edge of technology, drop a comment in to this article and let's see where we can take this.

TrackBack

TrackBack URL for this entry:
http://www.app-rising.com/cgi-bin/mt-tb.cgi/293

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)