IP Multimedia Subsystem (IMS) hit the telecoms industry like a lightning bolt three years ago. The new architecture, originally designed to deliver IP multimedia to wireless users, was seen as an inevitable standard that would bridge the transition from fixed to IP networks. Vendors such as Nortel, Avaya, Alcatel-Lucent and others stepped up their marketing of IMS, but many telecoms couldn't quite grasp the new business model and were reticent to embrace it.
Some service providers opted instead for the simpler service delivery platform (SDP) model, which involves a different approach to deploying converged multimedia applications individually such as VoIP, and video-on-demand and instant messaging on mobile phones.
'The term IMS is losing its luster in the market,' says Keith Nissen, an analyst for research firm In-Stat. 'It's now entering its deployment phase where it is being marketed as more of a service such as VoIP or video sharing. The term 'IMS' never comes up.'
Nissen notes that since IMS is really a network architecture, it's difficult to develop a viable business plan for it. 'From a technical standpoint, IMS is very important. But the reason IMS is not progressing as rapidly as many people thought it would is because the business model that is driving IMS deployment is not so well-defined yet. With SDPs, you are taking a step-by-step approach.'
As a result, Nissen and others say proponents are taking a much more pragmatic approach to marketing IMS, stressing the applications it can empower rather than the immediate complete network transformation it promises.
'The hype was very, very high on IMS,' notes Eric Bezille, a marketing leader for Nortel. 'Now it is more rational - do it when you need, if you need it. We still believe IMS is highly advantageous, especially for service providers interested in accelerating their time-to-market applications. However, in the past IMS was pushed to operators without a good justification for their businesses.'
Indeed, a recent survey by Comptel finds uncertainty about IMS worldwide. The poll, taken at the group's annual user group forum, revealed operators remain unclear about a migration path to IMS. The survey revealed that 32% of operators worldwide intend to move toward an IMS architecture within the next two years. Another 32% have not made any definite plans for IMS adoption. Another 27% expect deployment in two to four years, with another 6% looking beyond four years. The spot poll included operators from Asia Pacific, East Africa, Europe and the Americas.
Despite such skepticism about IMS, there are clear signs that the market remains strong. In-Stat estimates that subscribers to IMS-based services should increase from 10 million in 2007 to over 500 million by 2011, with spending on IMS control-layer equipment expected to hit $12 billion by 2011.
'There is no debate over whether IMS will be deployed,' says In-Stat's Nissen. 'Since all network equipment is becoming IMS-compliant, carriers will be deploying IMS regardless of their intentions. Mobile networks are evolving to IMS to reduce opex. Fixed-line operators, motivated by revenue retention, are deploying IMS-based point solutions on a very pragmatic basis, designed to meet each carrier's unique business goals. Whether service providers can adapt to meet the challenges of next-generation markets is a question still up for debate.'
A report from Ovum notes that Europe is a hotbed for IMS with early IMS deployments already being conducted by service providers such as BT, Orange, Telefonica, Swisscom and KPN. TeliaSonera and Telecom Italia are close to launching IMS initiatives. North America is also active. Asia has relatively small IMS implementation base at present, according to Ovum. The research firm pegs the current IMS market at $234 million with a compound annual growth rate estimated at 106%
'The business cases [for IMS] are around infrastructure consolidation and lowered opex for operators as they launch data and content services,' says Cynthia Leung, senior analyst for service platforms at Ovum APAC. 'In our view, an applications rather than an infrastructure focus, where the applications can act as a means for revenue-generating services to introduce better infrastructure, is more effective in attracting NSPs to invest in IMS.'
Dr. Abdul Razaque Memon, director, solution marketing for Alcatel-Lucent Asia Pacific, says there are more than 245 fixed and mobile IP/NGN networks worldwide ready to evolve to IMS. 'Currently, Alcatel-Lucent has 20 full IMS deployments in the field or in advanced trials,' he says.
These include carriers KPN, BT, Telstra and China Netcom. The Telstra implementations involve POTS emulation, where subscribers will not know if they are connecting to a traditional TDM exchange or a softswitch. The China Netcom implementation involves an NGN deployment 'in a couple of provinces across China, where all the POTs and broadband networks are connected through NGN networks and offering new broadband and multimedia services.'
Memon says APAC carriers are more judicious in their IMS approach. 'In APAC, we see that operators' IMS deployment stages vary, but they are moving at their own pace with many different insertion points. We are, however, encouraged by the interest in IMS from a broad range of operators in the marketplace.' He adds that APAC carrier implementations typically involve such applications as basic VoIP services, video sharing, presence and push-to-talk. 'They are looking for a global framework allowing the rapid rollout of next-generation services to increase ARPU,' he says.
Even so, Memon conceded that the pace of IMS acceptance 'is slower than many analysts had originally forecasted. This has led to longer certification intervals, which often delay the full-deployment ramp. We don't view this as a setback; it reiterates that in the short-term, adoption will move along on steady revenue growth, versus a huge and sudden ramp-up. Overall in the APAC region, IMS is better received in Japan, Korea, India and China in comparison to other countries like Australia, the Philippines and Malaysia.'
It's precisely that pragmatic approach that likely will guarantee IMS' ultimate acceptance and enable it to coexist with competing technologies such as SDP, many believe.
'SDP can be viewed as a valuable pre-cursor to IMS services,' says Ovum's Leung. 'What is required is the need to identify an 'IMS-ready' SDP that can exploit IMS once it is in place, while servicing today's need for fast and flexible service delivery. A well-placed SDP offers the need to impose a rigid infrastructure now, while providing a flexible backbone for existing, interim and future services.'
The need for SDPs helped IT companies such as IBM and HP gain an early foothold in the traditional telecoms space.
'If they have IMS services today, for example, and they can only deliver three IMS services and want to integrate these services into their web services"&brkbar; we take the SDP and help enable the service from the network and services from the web and integrate them together.'
Subbiah believes ultimately, SDPs and IMS will coexist in the NGN to the benefit of both. 'IMS allows us to deliver more network-based services, while SDP allows operators to deliver web services. But the world is going to be made up of both network services and web services,' he notes. 'Therefore, IMS and SDP are going to be complementary. I don't believe it's a hyped technology. It's just like any technology. It's not mature yet. Once it's mature it will be there. We've solved a lot of our problems.'
(Fiona Chau contributed to this story.)