On the Record with Peter Desnoes

Move over Akamai . There’s a new streaming media sheriff in town. Is iBeam Broadcasting a content delivery giant in the making?

Through a unique combination of satellites and fiber optics, iBeam Broadcasting has created a next generation network for the delivery of streaming media content.

As part of its plan, iBeam has enlisted big name investors like Intel and Microsoft in its streaming media crusade, while forming partnerships with leading ISPs like America Online and ExciteAtHome .

For now, though, Wall Street appears to be under enthused by iBeam’s game plan. At a recent $9 a share, the company’s stock price is trading slightly below is initial public offering price of $10 back in mid-May.

With revenue of only $3.4 million and losses of over $33 million last quarter alone, it’s not hard to see why. Before the April Net stock meltdown, this might have been a different story, but right now, story stocks are a tough sell.

We recently caught up with iBeam Broadcasting chief Peter Desnoes to learn more about iBeam’s long-term plans in its battle against Akamai.

ISR: Let’s start off by you giving us a brief overview of iBeam.

Desnoes: Sure. We are the premier or dominant streaming media network providing streaming media services to various content creators.

ISR: Okay, can we talk a little about the satellite and optics technology underlying your service?

Desnoes: Let me kind of explain or begin by saying- and forgive me for stating the obvious- but the Internet was never designed to do what we’re now asking it to do. So what does one need to do in order to permit streaming content – particularly broadband content, but even non-broadband content – to work on the Internet? Packet loss is a problem that is not getting any better. The average piece of content travels on average through four different networks. Seventeen to twenty three different routers handle it. There are public and private peering points that have congestion. Congestion on the Internet is death to streaming.

ISR: Yes. It makes for a jumbled mess!

Desnoes: UDP – the protocol for streaming – is one which does not easily permit the retrieval of lost packets and even if retrieved would probably not get there in time to allow the streaming media experience to function very effectively. That’s why we have jitters, break up and audio out of sync with video and re-buffering. It clearly doesn’t work. So what are the things that people need to do to overcome these problems? One of them is the creation of a video transport layer. That is the encapsulation of the packets of data so that if you can’t monitor the protocol for the streaming content itself, a video transport layer – that which iBeam has developed and is its core technology- enables us to manage and monitor content over our own network. It doesn’t matter whether that network is satellite or fiber.

ISR: So you’re really pushing content to the edge of the network and closest to each end user?

Desnoes: Well, that’s where I was going to next. But before you even get there. The majority of content is going to go over a traditional network. The ultimate objective is to be able to bypass the traditional Internet backbone. It will be some time before we get there. Right now, iBeam – to skip
ahead a bit – has contracts in place with access providers that give us between 25 and 40 percent of the Internet eyeballs that can be delivered by satellite from the edge. But even at that, that still means that somewhere between 65 and 70 percent of all the eyeballs are not going to be delivered from the edge by satellite.

ISR: Right. So what does that mean?

Desnoes: If the majority of traffic is still going to go over a traditional backbone to last mile providers of one kind of another, what can we do that can enab

le us to manage and monitor the distribution of this content? We must do it in such a way that we can redirect it to servers placed not necessarily closest to the end user. We can minimize disruption and minimize inefficiencies within the traditional network. That’s something that iBeam has created and is at our core in providing some differentiated capability to others. Other companies go and put a streaming server somewhere on its network and say – hey, I stream! But just because you can push streaming content into a fiber backbone, doesn’t mean that you have the ability to manager or monitor it.

ISR: Sure. Two entirely different animals.

Desnoes: So where was have focused, it took us about eighteen months as a company before we became generally available to content providers. The video transport layer that we developed is over a million lines of code and took forty man-years of development to create. That’s something that is a very material diffentiating element of iBeam versus others who claim that they offer streaming services.

ISR: So you’re saying this is very different than Akamai’s INTERVU?

Desnoes: Absolutely. I’m not trying to pick on Akamai, but there is a good example. Akamai has all of these capabilities – algorithms – that enable them to manage HTTP or HTML in a way that perhaps no one else can. It gives them no ability to do anything with streaming! They’d be the first to tell you that if you asked them. They would now tell you that they’d been working at if for six to nine months and they believe that they are now able to do some things for streaming that they will be offering to the world in the near future. I’m not saying that others can’t duplicate what we’re doing, just like what Akamai is doing, but absolutely what we do is very much to streaming what Akamai is to HTML.

ISR: So basically you’re saying they’re stuck in text based HTML land and you’ve got the next generation streaming equivalent?

Desnoes: Absolutely. Which is why if you look at what the top five customers were for INTERVU at the time it was acquired by Akamai, they are all iBeam customers. Why? Not because we’re better guys. It’s because we have offered the next-generation technology, which is the leading edge of streaming distribution.

News Around the Web