New SDP Architecture needed for 'The Long Tail'
Aepona have some strong and interesting views on SDP architecture. We asked Michael Crossey, who's a stimulus presenter at our Technology Insiders' SDP/IMS workshop at the Telco 2.0 event this Thursday to explain the impact of 'The Long Tail' on SDP architectures. He says:
"Those of you who are regular readers of Telco 2.0 blogs, and followers of developments in the world of Web 2.0 in general, will no doubt be familiar with the concept of the "Long Tail". For those who aren't, the Long Tail is a term used to describe certain business and economic models in which the accumulative sales value of niche or specialist products exceeds that of the most popular products. This effect tends to be associated with Internet-based business such as Amazon and iTunes, where the costs of inventory storage and distribution are low relative to traditional businesses.
For such business models, research has shown that consumers place much higher value on the availability of niche products that appeal to their specific interests, than on the pricing of those products. Increasingly, this concept is being applied to content such as video, particularly as telecommunications operators consider how they can differentiate emerging IPTV services against the current offerings from established mass-market broadcasters. Some of these operators are looking to leverage the Long Tail effect by offering a broad range of specialised IPTV content to their subscriber base, taking advantage of an IP-based storage, transport and distribution infrastructure that allows them to achieve this cost-effectively.
So what has all of this got to do with Service Delivery Platform architectures?
Put simply, SDPs will have to change dramatically to adapt to this new economic model. Up until now, SDPs have been built to cater for "classic" content distribution models, where a relatively narrow range of broad-appeal content, from a handful of content providers, is offered to the subscriber base. These SDPs were effectively large, bespoke Systems Integration projects containing a high number of proprietary products and interfaces, with the majority of the cost to operators being attributed to professional service fees. As a result, operators could only form relationships with a small number of content providers, since creating and maintaining these relationships is a manual, resource/time-intensive and ultimately costly process.
This has, in part, led to the emergence of content aggregators in the value chain. However, relationships with content aggregators involve the same issues as described above, but more significantly, having another player in the value chain clearly reduces potential revenues to both the telecoms operator and the content provider, as well as reducing the flexibility for the operator to fully exploit network capabilities other than simple messaging (e.g. call control, rich presence, profile, location, and so on).
To adapt to the Long Tail model, a new kind of Service Delivery Platform is needed, one that embraces recent advances in IT Technology and that is built around standards-based, cost-effective products, not costly, bespoke integration services. Key technologies for this new SDP include Service Oriented Architecture (SOA), Web Services, and Enterprise Service Bus. Most importantly, the new SDP should embody the concept of a "Loosely Coupled" architecture, whereby each party involved in the delivery chain brings unique capabilities or assets, coupled together via SOA to create a complete service in a cost-effective and highly scalable manner.
With this type of SDP, operators can partner with a much broader range of content and application providers, many of whom will be small (perhaps even single-person) players offering targeted, highly specialised services.
This is made possible by, in part, by Web Services, the de facto modern standard for a distributed service architecture. With a Web Services based infrastructure, an organisation (however small) can draw on a huge range of programming resources and need no longer rely on highly specialised engineers.
This is, of course, an imperative for the mass market, for which telecoms-specific interfaces such as SMPP are inappropriate. With a Web Services/ SOA based SDP, it is straightforward for an operator to compose the actual services that are used by the third parties. These services, whilst exposing the standard telecom web services, implement the necessary access control and business logic personalised for each partner.
A critical element of the new SDP is the provisioning and management of third party content and application providers, a function which addresses issues such as security, policy, SLA management and billing. In order to provide the required scalability to address the high number of partners envisaged under the Long Tail model, this needs to be a "light touch", highly automated function based on standard, widely understood protocols and interfaces. Moreover, this function needs to permit changes as services evolve, again in a light touch fashion.
This is where Web Services oriented technology comes into its own. Discovery and provisioning via Web Services allows the management and control to be integrated with an operator's more general account management.
In summary, the time has come for the industry to embrace a next-generation Service Delivery Platform, one that allows telecoms operators and content/application providers to take full advantage of the emerging Long Tail economic model, and that is based on modern IT concepts such as SOA and Web Services using standards-based products rather than costly, inflexible Systems Integration projects.
Several pioneering operators, such as TELUS in Canada, are already well down this path, using Aepona's Telecom Web Services platform to realise such an SDP. When this approach is more widely adopted operators globally, then the full potential of Telco 2.0 can be truly realised."
Come and join the brainstorm on Integrating Technical Architectures with Telco 2.0 Commercial Realities this Thurs. Register here.