|
International Issue: June 2005
Can you manage without next-generation OSS?
Service providers want both a standardised IP network and differentiated end-products. Next-generation operations support systems (OSS) need to bridge the divide
by Ouida Taaffe
The marketing category that is breakfast cereals is, arguably, a triumph of chutzpah. You can, for example, have corn flakes, corn biscuits, corn puffs, crispy corn puffs, crispy hexagons, sugar-coated octagons, and a dozen other variations on what is, at bottom, a not overly differentiated product: corn. Corn in the raw, of course, doesn’t really command much more than commodity prices. Still, cereal manufacturers have been undaunted and crispy puffs, if properly marketed, can — over time — command good margins. The question is: can the operators of commoditised IP networks pull off a similar trick? The networks themselves, or at least their management systems, could prove a stumbling block.
Build once, twice, a thousand times
Legacy operations support systems “historically grew like moss over stones”, says Paul Gowans, business development manager for EMEA at Agilent, which has OSS network and revenue and management solutions within a broad-based test and measurement portfolio. “It used to be ‘if I roll out a new service, I need a new OSS for that’,” he says, adding that some major telcos can have thousands of OSSs in place — many of which were set up in splendid isolation. In a monopoly environment, this was not a problem.
“In the past, different networks addressed different needs. Having distinctive networks was the only way to get the job done and they could pass the cost along, pretty much unhindered, to the consumer,” points out Glen Tindall, CTO of Intelliden, a US-based provider of network management software. However, in a world where, at least in theory, flexible service provisioning will be part of what differentiates a provider and keeps margins healthy, the “stovepipe” approach could hamper good business practice.
“For a next-generation network not to become a replica of the challenges we see in present-day networks, the network management side of it has to be treated in parallel and with equal importance,” argues Martin Creaner, CTO of the TeleManagement Forum (TMF). “If that happens, we will see a very different telecoms world in five years’ time. If it doesn’t, I don’t think we will see much difference from the consumer point of view.”
There is, some argue, an even more radical take on the move to next-generation OSS. “There is no point in rolling out broadband and IP networks unless their OSS can support the services,” says Sandy Aitken, partner at IBM Business Consulting Services.
Operators are somewhat cagey about discussing their OSS strategy. However, there does seem to be general consensus in the wider market about the wish list. “Service providers want four main things from next-generation OSS: opex reduction through automated service fulfillment, ‘service velocity’, service assurance and ROI,” says Verne Anton, an OSS analyst with Gartner. There are vendors who envisage an even tighter focus. “The economic rationale for the next-generation network it the ability to integrate all the service on one platform and the same logic applies at the OSS level,” says Don Gibson, CTO of Cramer. “They all want to reduce their opex,” says Gowans. “The way a lot of carriers look at it is that a capex budget is a one-off, whereas opex is for life.”
It could be argued — as, with reference to cuts in equipment expenditure, a report from Gartner recently did — that shrinking prices in one aspect of the network leaves a vacuum that could be filled by ballooning costs in others — say, with more outlay on software and integration. Integration is already a large and constant headache in present-day OSS.
“The business cases I’ve seen so far come out positive,” says Aitken, who was reluctant to give details on whether or not next-generation OSS implementations actually pay off. “But it does mean that you have to embrace the change. What happens quite often is that people do part of it. And that doesn’t solve your problem.” Aitken argues that the implementation of next-generation OSS represents a transformation of the businesses involved. “I know lots of consultants say that,” he adds, “but it is a watershed in telecom.” Anton, at Gartner, suspects that some new costs could appear. “However, there are significant cost pressures on providers, so overall spending will go down. Off-shore systems development will help that,” he says.
A tax on both your houses
What sort of money could service providers save? Gartner estimates that the global spend on OSS alone in 2005 will be around US$18.6 bn. This number does not include what is quaintly known as the integration tax. For every one dollar spent on OSS software, according to the TeleManagement Forum, US$1-4 more is spent on integrating that software into an existing environment. “This isn’t because the products are badly designed, it is just that there haven’t been sufficiently comprehensive standards on integrating, on interfaces between various components,” says Greaner. This spend on OSS and integration, of course, it also not the end of the story. There is no point in managing services that you cannot bill for. The total annual global spend on OSS and BSS comes out at around US$30-40 bn, according to the TMF, again without the integration tax.
The TMF set about tackling the integration tax issue with its Next Generation Operations Systems and Software (NGOSS) programme at the end of 2000. There are, Creaner says, three main strands to this. The first is the definition of interfaces between software components. One recent development in the field of software interfaces is MTOSI — the Multi-Technology Operations Systems Interface, a standard OS interface supported by Cramer, Lucent, Nortel, Siemens and Telcordia that was announced at the TeleManagement Forum in Nice in May this year. The stated aim of MTOSI is to “encourage application inter-working, faster deployment and lower cost of ownership for next-generation operations systems”.
The second main strand in NGOSS is the definition of a common data model for the industry — something that Creaner sees has “probably the biggest challenge. People have been trying to do that for about thirty years”. As things stand data models are far from standard. For example, one application need not define or classify information on, says, customers in the same way as another.
The third area is defining a common set of business processes for the telecom industry. As things stand, processes like service provisioning are more or less the same across the industry. The TMF aims to remove the “more or less” aspect of this and nail it down with a unified approach. Creaner says that NGOSS has “been widely accepted and adopted”. The Shared Information and Data model (SID) and common business process — Telecoms Operations Map — eTOM — have been defined and eTOM is being adopted by the ITU as the industry standard.
“Service providers want it, the vendors want it. You may question whether the people who make money out of this integration, the systems integrators (SIs), want it,” says Creaner. “However, I think the SIs see the writing on the wall and are becoming the people who make it go away and move up the value chain.”
NGOSS is not the final word on next-generation OSS. The TMF is currently working through a gap analysis, in concert with organisations including the European Telecommunications Standards Institute (ETSI), the ITU and the Third Generation Partnership Project (3GPP — an umbrella standards body for third generation mobile). Creaner expects a meeting planned for the end of July to give some clarity “in terms of what works”. He sees a lot of the current NGOSS as applicable, even though “we will probably need to tweak some bits or expand some bits”. The essential shift, in his view, is from a network centric management model to one where the customer and money are key.
What should it be able to do?
What, however, apart from cutting costs, should the practical gains of a new approach to OSS be? “A zero-touch, complete flow-through environment, with almost no human intervention,” says Aitken. “I think it is doable. It is about how you interact with customers and telecom is moving now in the footsteps of large businesses like banks.” What the TeleManagement Forum would like to see is an integrated approach to back-office management that ensure that the “little silos of capability” that bedevil networks today do not develop in the future — at least not for architectural reasons. Creaner admits that warring operational fiefdoms do develop within organisations, even small ones, and that these can mean co-operation is hindered for other than reasons of network management. (The TMF is not, however, out to try to change human nature.)
The key issue for next-generation service providers appears to be standing out in a market where the underlying network is a uniform given. “Everyone will end up with a core IP network,” says Yogen Patel, vice president of product marketing at Ceon Corp, a US-based provider of fulfilment and service management software. “The question is, how do you provision quickly and accurately and give customers control of their services?” As Tindal points out, having a broader service bundle ties customers in, and makes it harder to churn. Better OSS also makes it easier to get customers in the first place. “One DSL proivider in the Americas was taking 30 days to activate subscribers. We reduced that to same-day flow-through,” says David Sharpley, vice president of marketing and product management at Metasolv, a service fulfilment software provider.
 Spending on OSS over the next 5-6 years will see single digit growth — it is around six per cent this year. That won’t support the number of vendors that is currently out there.
Verne Anton, Gartner
No more OSS?
If next-generation OSS is about making service provision highly responsive, about, as Creaner says, the customer and money, rather than the internal functioning of the network, what impact will it have on its customer-facing cousin business support systems (BSS)? “There is a blurring of the division between OSS and BSS, there is a need to take a three hundred and sixty-degree view of the consumer into the network,” says Birger Thorburn, director of the global content practice at CSG Systems, a billing provider. He believes that the main shift will be towards a need for the network to talk to the system that manages the customer finances. The traditional division that labels OSS as network-related and BSS as an IT function, will, according to Thorburn have to give way to a more integrated process. “Sometimes, when we talk about converged solutions one of the biggest challenges is within the operations,” says Thorburn.
Presuming that the industry does move toward a standards-based, flexible OSS, with closer tie-ins to BSS, what will the impact on the OSS industry itself be? “Spending on OSS over the next 5-6 years will see single digit growth — it is around six per cent this year. That won’t support the number of vendors that is currently out there,” says Anton, at Gartner. “Though that doesn’t mean that new players won’t appear.” Anton also believes that some companies that are currently in the background of the wider OSS business, such as Oracle, could move further into the mainstream in roles such as inventory management. “Functions like network management, however, are likely to remain the purview of those with fundamental network experience,” he cautions.
The operator consolidation currently being seen in the market could also be a longer-term issue for OSS vendors. With fewer players, OSS could, arguably, have fewer sales outlets — although the network integration work thrown up by M&A obviously provides good pickings initially. Merging two back-office systems is regularly cited as one of the biggest problems that telecom mergers throw up.
“Clearly, there will be some consolidation,” says Sharpley, of OSS overall. No-one, however, seems to expect the emergence of a mega-OSS provider, in the image of big equipment vendors that are many things to most providers. “I just don’t think operators would rely on one single vendor to cover the entire scope. The risk would be too great,” says Sharpley.
|
|
|
