Content delivery networks (CDNs) are hardly new; they've been a part of the Internet's content food chain since...
the 1990s. A CDN is simply a ring of content storage points that are located near access providers and used to make content experiences both better and cheaper. The basic principle behind the original CDNs was that content delivery would be more reliable, and the experience better, if the content source was closer to the recipient, normally in a temporary storage mechanism called a "cache."
In the commercial CDN market that has evolved since then (with Akamai still the market leader), CDN providers are paid for caching content for delivery. But deeper CDN value propositions are shaping operators' CDN plans.
Using content delivery networks to monetize video content
Once envisioned as free on the Internet, video content has become a larger part of paid content models, as evidenced by companies like Netflix and Hulu that offer paid content streaming (usually long TV episodes or movies).
Enlightened design and sharing of [content delivery networks'] resources could improve customer experiences and more effectively harness resources.
Tom Nolle, president, CIMI Corp.
Video content represents a significant and continually growing traffic source. The additional traffic has already created tension between traditional CDN providers and Internet service providers (ISPs) that want to be paid for handling the transit traffic that comes with video, rather than allowing it to flow freely under existing peering agreements.
But Internet peering and transit payments don’t provide much revenue, and ISPs have long seen CDNs as a way of charging content providers for a caching service that helps the operator pay for traffic generated by video. As those operators looked deeper, they realized that CDNs would likely be an element in developing specific strategies to monetize video content, as well as a tool for managing capacity utilization within their networks.
Simply streaming online video in competition with networks or movie services is less attractive than creating social viewing frameworks to share video, supporting the easy migration of a viewing experience from one device to another, or providing customized video metadata to help viewers find the right programming and even link related scenes in multiple videos. Operators are experimenting with all of these approaches, and nearly all cases involve CDNs.
How content delivery networks can improve mobile user experience
Mobility also creates CDN opportunities because mobile services are delivered over network connections that are not only constrained in terms of capacity, but have varying capacity depending on the customer's location relative to the cell tower.
Mobile services are also subject to dynamic changes in the delivery route as the user roams between cells. When the user roams into another network, the normal process of homing traffic can create long and wasteful video routes, popularly called "hairpinning" or "tromboning." Enlightened design and sharing of CDN resources could improve customer experiences and more effectively harness resources.
More capable CDNs needed to leverage video, mobile opportunities
To achieve these goals, operators are looking for a new set of capabilities found in only a limited way (if at all) in traditional CDNs, including:
- Support for a highly flexible strategy for cache location, not only to sites at the edge of the access network, but also within the metro/access network itself.
- Support for the integration of mobility and backhaul considerations into the decision on what cache to draw information from, and for the dynamic switching of cache locations when the user has moved between cells or even between operators.
- Provisions to allow users to switch screens during content viewing, to share content views with others and to integrate content viewing with parallel social network or SMS interactions.
- Provisions for integrating advertising into content in new and interesting ways, supporting ad targeting in a way that is less likely to create regulatory scrutiny and consumer backlash.
- Facilities to transcode video to a format suitable to the type of device, conform to the limits of the delivery path, accommodate a change of devices or select a different cache with a different format, should any of these variables change.
- Integration of video metadata describing the content theme, scene structure, etc., to support easy locating of video and jumping from one video to another based on thematic interests.
Perhaps the biggest operator issue, and the place where operator CDN models are most likely to differ from traditional models, is the area of federation. Operators generally plan CDN deployments in their own operating geographies, but mobile services and partnerships between channelized video players and IP delivery players create a need to share CDN assets to create a broader service footprint. As they develop, these mechanisms may then become the same ones used for the federation of broader service-layer components and features. That could create a whole new model of the global network.
Nearly all of the Tier 1 and Tier 2 telcos and cable multiple system operators (MSOs) are exploring their options for the creation and deployment of CDN services, and equipment vendors are assembling and packaging their assets to support these new missions. As these trials and tests evolve, it's becoming clearer that the priority of network operators is to link CDNs with content monetization plans and strategies for optimizing content delivery's return on infrastructure, rather than to support a "traditional" CDN mission. That means that network operator plans for CDNs may well revolutionize not only the CDN market, but also the online video experience.
About the author:
Tom Nolle is president of CIMI Corporation, a strategic consulting firm specializing in telecommunications and data communications since 1982. He is the publisher of Netwatcher, a journal addressing advanced telecommunications strategy issues. Check out his SearchTelecom.com networking blog, Uncommon Wisdom.