P2P Delivery Networks Can't Survive On Their Own

Since the acquisition of RedSwoosh by Akamai, a lot of P2P delivery providers have been contacting me saying that the acquisition now validates their service offering. One has nothing to do with the other. Over the years, we have seen many acquisitions in the market where the company that was acquired then got close down or went under. The fact that one company in the P2P distribution market got acquired does not mean that all of a sudden P2P is going to get adopted.

Yes, it has a chance to become another means of distributing content, but it's not about comparing P2P distribution to other forms of content distribution. So many people want to talk about P2P as if you have to use that, or something else and no one talks about it as a hybrid. It's like the people who say they will either do streaming or progressive downloads. It's not one or the other. It's always about using the best combination of delivery platforms depending on the type of content being delivered along with who it's being delivered to and how they want to consume it.

In order for P2P to make it, it has to be combined into a service based business. It can't stand alone and be supported by itself. P2P can't try to fight against the traditional means of CDN distribution or else it will lose. It has to be combined with all other forms of delivery and be thought of as just another means of video distribution.

P2P vendors also need to stop referring to Joost and BitTorrent as successful P2P companies and trying to use those as examples when talking to people about P2P. Joost and BitTorrent are NOT P2P companies. They are content companies who happen to use P2P for delivery. But if consumers do not adopt Joost or BitTorrent's content business models, then the form of distribution they are using does not matter. They are content companies trying to sell content, that's the bottom line. The consumer does not care how the content is being delivered or what the protocol is as long as it meets their own personal expectations of how they judge quality.

It's a natural fit for a traditional based CDN like Akamai to want to add P2P delivery to the many other forms of content delivery offered and I expect we will see more CDN do this over time. That is the best way to get P2P adopted, offering and using it as another means of delivery be it streaming, progressive download or P2P.

Comments

[Marco Parente]

Dan, we have received inquiries from P2P only providers looking to partner with a CDN to ensure QoS. Some of the challenges these companies face is finding a CDN that understands P2P and knows how to integrate with their P2P offerings. Most want to avoid the alternative of having to build out and support their own global infrastructures so they can focus on their core business.

[Steve Lerner]

P2P has been tried in various markets for a long time. Anyone remember short wave radio (ham radio)? Or CBs?

P2P software for data transfer is not much different- an economy of scale and cost can be gained if the end-user takes on some of the cost and load for a system.

P2P can be integetrated with a CDN to add quality of service- and is an elegant solution to a few problems of moving large files.

I'll be interested to see how the delivery platform of the future evolves.

-Steve Lerner
Practice Leader
RampRate Sourcing Advisors / Strategic Research

[Christopher Levy]

There is something to be said about the efficiences P2P technology affords that are hard to overlook.

Conversely the QoS issues, concerns about security of content flowing through P2P and the end-user machine impact have weight.

On one hand CDN's can benefit from offloading some of their cycles to P2P clients but in doing so they are passing off this load to the end-user which may not be the experience they seek.

We have all watched the CDN's evolve in the past 8 years and the marketplace will naturally drive where technology is applied. Today end-users may not want P2P but tomorrow, for a compelling experience that may include lower costs, more content and "closer" content, users may be willing to go along for the ride.

[Krys Cameron]

Is there anyone close to providing a P2P solution for content providers like myself over my network site?

Akamai is saying that they don't offer such a solution, and do not see doing it at all.

[Yaron Samid]

Hey Krys,

Email me at yaron@pando.com and we'll get you set up.

-Yaron
www.pando.com/channels

[Yaron Samid]

Dan - I couldn't agree with you more (and I've tried on several occasions). P2P will reach true mainstream ubiquity as one of several channels integrated into comprehensive media distribution platforms. We are now in the early, exciting stage of initial consolidation and standardization.

At Pando Networks we've architected from day one to be embeddable in native media workflow platforms and applications. Our flagship consumer app and upcoming publisher tools showcase our managed p2p platform, but the future is all about seamless integrations with leading media publishing, distribution and monetization platforms. P2P, indeed, should not stand alone - its far to valuable to the future of the Internet and its outdated infrastructure.

[critster]

I disagree! (respectfully)
There is a huge emerging content distribution market, requiring a variety of content distribution models, and there is indeed room for for P2P delivery networks to survive on their own.

The underlying technology is advancing at a rapid rate. QOS, QOE, and network efficiency with the latest technologies are starting to catch or surpass people like Limelight. That being said, you do need a new vocabulary of terms like "contribution ratio" and "convergence time" if you want to properly evaluate respective technical approaches.

Clearly, building a content portal simply to drive footprint for your P2P solution is a misguided approach if your intent is to be a technology vendor.