Aiming to unclog legitimate P2P

High-definition movie files and large software downloads could find their way more quickly to end users, if a new platform takes off.

Velocix is described by vendor CacheLogic as a multiprotocol P2P-based media delivery platform that could revolutionize legitimate peer-to-peer traffic. Velocix is a system of local caching, whereby popular high-volume files are stored at various locations relatively close to where they are needed.

The idea is that agreements could be signed with content distributors, such as movie studios or software vendors, and Internet service providers, enabling large files to be stored with Velocix and distributed locally when requested. For very large files, this could cut transmission costs for the ISPs significantly. This could also encourage customers to download copyright-protected content rather than unlicensed versions.

However, no such agreements have been announced yet by CacheLogic.

"The content world has become increasingly interested in P2P from a legitimate distributive perspective," CacheLogic's chief technology officer, Andrew Parker, told ZDNet UK on Friday.

He claimed the primary reason for this was cost reduction, as P2P "creates economics that are far closer to broadcast" than streaming video over the Internet.

The current problem is that content distributors such as Warner Bros. or Universal sell, for example, a television show at a price not much higher than they charge for a music track. Because a television show--especially if it is in high definition--is much larger than a music track, the customer's ISP has to pay much more to transmit it.

"This is one of the things that has really held back the delivery of video over the Internet," said Parker, who asserts that Velocix provides a "hybrid of traditional P2P systems and content distribution networks, offering the scalability and cost benefit of P2P delivery but with the reliability and consistency of a content distribution network."

By enabling ISPs to "keep temporary copies of the most frequently requested content on their network," Velocix could possibly make downloads faster than if they came directly from the content distributors themselves, but it remains unclear whether there would be any speed benefit over other P2P services.

Content distributors could also be reluctant to take up such a service until they recognize the impact of ISP capacity limitations, as "a lot of them are very new to this kind of P2P content distribution and delivery market," according to Ovum analyst Jonathan Arber.

"I think, for the ISPs, it's very attractive. But my feeling at the moment is perhaps the content owners will look at it and say it's a nice idea but why do we need to bother," Arber said. "At the moment, their services are only seeing very slow uptake. Mass-market usage will create the problems that this solution is designed to remove."

Arber added that a service such as Velocix could theoretically enable ISPs to further segment their services, by "prioritizing packages to high-usage users who are using huge amounts of media content."

It is also thought that the service would be useful for software distributions and large security patches, such as the Centrino patch issued by Intel last week.

Several open-source companies already use the BitTorrent P2P service to distribute their packages around the Web, including OpenOffice, Suse Linux and Ubuntu. This lets a person download a single file from many different places at the same time, which speeds up the process.

David Meyer of ZDNet UK reported from London.

[Jackson Cracker]

is this really peer-to-peer?

The type of content is different, but this sounds more like what Akamai is already doing.

[Penguinisto]

Nice, but with a couple of caveats:

1) Is DRM going to be enforced? (it'll be a PITA to play paid-for content if one doesn't have the OS/player-software/codec required by the DRM mechanism)

2) Is it just for pay-to-download files, or can free/OSS content get in on some of this as well?

If it's DRM-only and pay-only, I want no part of it. Otherwise, it would be useful indeed.