March 10, 2011
By Dom Robinson Contributing Editor
Featured Articles
For the rest of the Spring 2011 (Sourcebook) issue of Streaming Media magazine please click here

Telcos and the State of Content Delivery

This means that while the CDNs have the buying power to drive peering and even transit costs down, they do not have the vast infrastructure overhead associated with creating the demand for the traffic. Accordingly, they can make a greater margin at lower internal cost to their clients than the TDN.

That said, the TDNs reach physically closer to the end users (as mentioned, often they span the entire Tier 1 through to local exchange and end user). They can include caching at the local exchanges, somewhere that CDNs struggle to reach, since there are so many telcos globally to deal with and, usually, only two or three in any area, all of which fiercely defend control of those exchanges.

By including caching and splitting in the exchanges, the TDNs can dramatically reduce the amount of content that they need to deliver across their own networks, though all the while, they continue to charge the same at the transit and delivery side. So they reduce their internal cost while maintaining their revenue and increasing their margins—always from a strong position when it comes to the competing CDNs, since the CDNs will generally still need to reach the end users over the telcos’ access networks (the ISPs); thus the telcos can charge the CDNs for transit. In reality, the telcos don’t generally charge for access to their access networks—that is paid for by the ISP. However, the CDNs are “charged” in terms of their network priority. If the telcos want to, they simply deprioritise the CDNs’ traffic, and the end users experience poor delivery. This gives the telcos the ability to prioritise their own traffic and charge a premium for this priority. This is the very heart of the Net Neutrality debate that is reaching heated proportions as this article is being written. It will continue to be a key policy discussion and a key technical issue during 2011.

Video and the TDN

IPTV models exist within specific telco networks, so the quality can be guaranteed just within that network footprint. But you have to have a physical connection to the telco providing the IPTV service, and this has to pass by your front door for you to be able to engage with it.

The big advantage with the internet is that any connection gets you onto the network. This is as equally true for publishers as it is for subscribers.

In Jet-Stream’s excellent “Telco CDN Strategy White Paper” (http://jet-stream.nl/blog/2010/09/the-telco-cdn-strategy-white-paper) released last year, CEO Stef van der Ziel states that “consumers want the freedom of the internet but with the QoS of walled garden services. With On-Net CDNs, telcos can deliver that promise.” Indeed, the Jet-Stream product set is ideally suited to assisting in meeting that demand, and judging by Jet-Stream’s growth these past 3 years, the demand is significant.

So if your content model is about getting the content delivered because you want the content “out there” anywhere possible, then the CDN model is a very good approach. If you want optimal quality within a specified region, then the TDN model has many advantages over the CDN.

I personally see the “stack creep” (see www.streamingmediaglobal.com/Articles/Editorial/Featured-Articles/The-State-of-the-Stack-69563.aspx) as a key effect here. If the stack model is accurate, then the CDN and the TDN are “applications” running on the networks. The CDN is really a caching application for moving data between autonomous systems to lower cost and optimise quality. The TDN model is a caching application for moving data within discrete collections of autonomous systems that it controls, and this increases the reutilisation and availability of these networks, although it arguably probably does little for quality, since the networks are quality- controlled by a single entity from the outset.

The nuances are hard to firmly define, but they are clearly identifiable. However, it remains to be seen which of the TDN and CDN is the wasp and which is the bee.

This article originally ran in the 2011 Streaming Media Industry Sourcebook as "New Year, New Acronym: Telcos and the State of Content Delivery."