Pando CEO: The 'CDN of the Future' Will Include P2P

Source: http://www.contentinople.com/author.asp?section_id=450&doc_id=160788
4 Aug, 2008
Short Summary: Pando Networks Inc. CEO Robert Levitan says the dirty little secret about online video is that "the business model is really bad." He attributes this to today's current delivery model where, he says, "the more video you deliver, the more money you lose."
Intercast Feedback:
6 Aug, 2008
Author: Alon Levitan, Intercast Networks AVP Marketing
We all know why the rise of P2P technologies is stalling. Many of the leading service providers (BBC is a good example) understand that P2P cannot deliver the quality of service levels to provide a continuously good user experience. No larger brand is going to jeopardize its brand equity to be associated with a quality prone product or service.

Being dependant on the amount of users that are concurrently online, that have opted in to seed a specific title (video) and that are not using their already narrow upstream bandwidth for other purposes is simply too much to ask for.

Therefore, and not surprisingly, most of the P2P network providers must revert back to streaming technologies to make up for the lack in P2P delivery efficiency.

Is P2P going to be used in the future to replace a portion of the CDNs delivery over internet? I suppose it depends mainly on the ISPs to embrace this idea. With oversubscription as their main business model and frightening and fast increase in bandwidth consumption P2P only adds to the over-the-top traffic problem that ISPs are trying to cope with.

Network providers and ISPs have other delivery technologies they can choose from. Why not utilize the entire bandwidth spectrum at hand. Off-peak time bandwidth is available in abundance and is just waiting to be tapped into. What if there was a way to shift the majority of bandwidth intense traffic - yes, we are referring to video - to the most economical time window/s on the network. All this underutilized bandwidth would suddenly come in very handy and improve each network's/ISP's overall delivery efficiency and of course bottom line results.

What will it take for the network players to understand that the solution to their problem lies in the activation of IP multicast on their networks? If we then add a personalization component to the equation, which enables to determine each user’s personal preference and then push the content directly to each user’s personal storage via a scheduled delivery process, we suddenly get a very smart, efficient and scalable network that can handle the onslaught of video traffic. Multicast-to-Storage technology delivers on this promise and therefore needs to be considered when looking at the future of content delivery networks.
In the end the future of content delivery will be a mesh up of different technologies each providing a crucial component in the delivery process. The user will not care how he receives his personalized content; he just wants to get it in the best possible quality.

File-sharers want to have your cake and eat it too

Source: href="http://www.guardian.co.uk/technology/2008/jun/19/digitalvideo.internet
19 Jun 2008
Short Summary:
Peer-to-peer file sharers think it's perfectly OK to grab three quarters of the communal internet bandwidth. Indeed, some are defiant about it. Their internet service provider has foolishly sold them an "unlimited" connection so they are entitled to download 5GB a day, or more, at any time. The fact that today's internet is incapable of coping with their demands is beside the point: ISPs should simply provide more bandwidth.

Intercast Feedback:
19 Jun 2008
Author: Noam Bardin, Intercast Networks CEO
The need to migrate from the unicast internet to the multicast internet to support the evolution of the web from a communication network to an entertainment network. Another excellent article on the limitations of the unicast internet and a proposal to limit demand for online video delivery technologically (i.e. since it is expensive to deliver what people want, lets limit what they get...). At Intercast Networks, we believe that the source of the problem is the unicast basis of the internet (point-to-point delivery) which is great for communication and lightweight content but not optimal for video and heavy lifting of large files delivered to many users. The next evolution of the internet will have to be the migration from the unicast Internet to a multicast Internet (point-to-multipoint) which can replicate the economic structure of the broadcast world and is the underlying infrastructure of IPTV deployments while still supporting the innovation of the world wide web.

Cisco Visual Networking Index Projects Global IP Traffic to Reach Over Half a Zettabyte(1) in Next Four Years

Source: href="http://newsroom.cisco.com/dlls/2008/prod_061608b.html
16 Jun 2008
Short Summary:
Cisco Systems Inc. is projecting a sixfold jump in Internet traffic between 2007 and 2012, as online video becomes the biggest driver of global data communications. The networking-equipment maker, as part of a study called the Cisco Visual Networking Index, predicts that Internet video - which accounted for 5% of data traffic in 2005 - will represent 30% of total data transfers by the end of this year. That will swell to 50% by 2012, Cisco estimates.

Intercast Feedback:
16 Jun 2008
Author:
Exellent Whitepaper providing in depth details on the growth and usage of IP networking worldwide. Findings show that bandwidth usage will most likely double every two years mostly as a result of video sharing.