The quest for end-to-end telecommunications solutions is now achievable because of the widespread adoption of standards, says William F. Wilbert
Seems like everyone is searching for end-to-end telecommunications solutions that can be quickly and cost-effectively implemented. You'd think that business -- or at least offering a host of new services -- depended on it. And you'd be right.
The race to deliver new products ahead of the competition is putting enormous pressure on all carriers to move away from a proprietary world of inflexible, monolithic systems toward a much more open world of containers and components. Single vendor solutions have given way to multi-vendor solutions that include products from ISVs, Network Equipment Providers (NEPs), system integrators and middleware vendors.
The more compatible these products are, the easier they are to integrate and the better off everyone will be. The challenge is to build end-to-end solutions that support the full spectrum of OSS requirements. These solutions must also support increasingly heavy burdens on network management. Scalability, availability, reliability, and agility will remain top concerns.
Trapped between the old and the new
Though everyone agrees that legacy infrastructures are technology dinosaurs that will eventually be replaced, most carriers still depend on a tangle of processes, systems and networks -- a patchwork of legacy OSS/BSS and individually integrated applications, either homegrown or commercial off-the-shelf (COTS) applications.
Many believe that this uneasy co-existence will continue for some time. After all, OSS/BSS can't be unplugged without halting business, and opportunities for green field implementations are increasingly rare. But the pressure for change is mounting. Within the last few years, the complexity of OSS/BSS has grown exponentially -- by a factor of 10 some say, including Telstra, a leading Australian telecommunications and information services company.
According to Michael Lawrey, Head of Network Services Infrastructure Services at Telstra, the best way to deal with complexity is to step back and look at the problem from the 30,000 foot level. Lawrey, who also sits on the board of the TeleManagement Forum, strongly believes that telcos must start with business strategy, then look for standard integration platforms.
"If you want to solve the OSS/BSS problem, you've got to have an overall business model first before you can understand what the infrastructure building blocks should look like," he says. "Telstra is a legacy telco. It's grown up in its siloes. Within my business area I'm running a massive integration program. I've got all the fixed networks, the mobile networks, all the cable networks, all the online broadband networks -- all under my one business unit and they have to be able to function together."
To achieve this, Lawrey is striving for what he describes as "a fully integrated business model."
"You've got to start with a business model point of view and a process point of view," he says. "My real challenge is trying to get our systems to come online or drop off as needed. To succeed, I want company-wide standards, I want a framework. The big hurdle for us and for other telcos is to agree on a company-wide business model. Once we do this we will need some glue to hold the framework together. This will allow us to move from a siloed world into a new world of fully integrated services."
Standards are the strategy
To break the bonds of technology and process siloes and find the Holy Grail of end-to-end solutions, carriers are thinking long and hard about standards.
"Standards are necessarily strategic," says Peter Mottishaw, Strategic Marketing Manager, OSS Business Unit, Agilent Technologies, a major provider of service and network assurance OSS software. "You may get a more immediate benefit from a proprietary solution, but over the long haul you are likely to discover hidden costs if you go down that route. In fact, the hidden costs can be huge. This is why carriers are looking for solutions that plug and play together and can also be integrated into the OSS infrastructure fairly easily."
On the technical side, much of this boils down to integration issues. Tomorrow's infrastructures are being built now and depend on new ways to link existing systems and networks with new ones. After years of learning through painful trial-and-error experiences, the telecommunications industry has come to the collective conclusion that the best way to bring order to the middleware/application interface chaos is to adhere to standards.
Along with a host of other carriers across the globe, Telstra is now taking a close look at OSS/J, the industry's first implementation of the TeleManagement Forum's New Generation Operations System and Software (NGOSS) framework.
"OSS/J is the glue that holds system components together," Telstra's Lawrey explains. "I want OSS/J to put the glue in place to get all our OSS building blocks to talk to each other, to get my end-to-end process flow right and to get my end-to-end service management layer right."
The goal of the OSS through Java Initiative is to create a carrier grade industry standard integration platform. An industry-wide consortium of leading industry players-equipment providers, OSS suppliers, applications developers, and consulting services providers -- the group develops, tests and publishes inter-OSS application program interfaces (APIs) that are available free of charge. (For a full list of current members, APIs and certified products, go to www.ossj.org.)
OSS/J APIs are based on the J2EE (Java 2 Platform, Enterprise Edition ) integrated application framework. J2EE was chosen because members agree that it is "the simplest and most reliable" means of embracing a multitier architecture based on re-usable components and container technology, that can address tightly coupled integration with Java, loosely coupled integration with XML and Web Services. OSS/J deliverables are backed by best practices and design patterns implemented in each API. OSS/J's careful mapping of its APIs to the TeleManagement Forum's widely adopted enhanced Telecom Operations Map (eTOM) means that providers and vendors can apply common process integration rules to the integration tasks that comprise business process engineering and re-engineering.
They can also use common, freely available implementation tools to put those rules into practice. Endorsed by a large majority of the world's leading service providers, eTOM defines a common terminology and map for communications industry business processes and a reference point for internal process Â© reengineering needs, partnerships, alliances and general working agreements with other providers.
While OSS/J implements the fundamental architectural patterns common to all players within the telco industry, it has characteristics that set it apart:
* OSS/J is the only open standard for OSS that is available with reference implementations, compatibility test suites, a portfolio of certified products, and an open ecosystem of tools, adapters and extensions.
* OSS/J is the only open standard that builds upon a middleware that is itself an open standard: J2EE. Compatible implementations of J2EE standard are available from a large choice of suppliers.
* OSS/J implements telecommunication application open standards and runs them on open standard middleware.
"OSS/J adds OSS functionality to the standard enterprise Java and Web Services platform called J2E," says Philippe Lalande, Sun Microsystems, head of the OSS through Java Initiative. "A lot of enterprise software experience has been engineered into J2EE and telecommunications is now able to build service oriented architectures upon standardised OSS/J interfaces and mainstream enterprise platforms for their next generation OSS systems. Because OSS/J certified products can be assembled in a plug-and-play manner, OSS/J has become the implementation standard of choice for the successful and rapid deployment of next generation OSS solutions."
Agilent's Mottishaw agrees with this assertion. "It's clear from our perspective that the J2EE and OSS/J environment have become well established in the industry," he says. "Within our customer base, J2EE is increasingly being used as the platform for integration. J2EE is the preferred integration environment and OSS/J APIs are a critical part of this. So I think the role of OSS/J within the industry is pretty well assured. There's really no other option right now."
Independent industry analysts confirm the fact of widespread OSS/J adoption. "The battle over which standard will dominate is essentially over," says Dan Baker, director of OSS research at Dittberner & Associates. "The J2EE environment is the clear winner and the new OSS/J tools basically let you write from an expert outline instead of from a blank piece of paper."
J2EE (together with its telecommunications extension OSS/J) has proven that it can facilitate actual end-to-end software solutions that support the full spectrum of OSS requirements.
OSS/J and Open Source
There are multiple indicators showing that the pace of OSS/J adoption has increased markedly over the last year. Among the service providers who are adopting OSS/J, BT, Covad Communications, Vodafone and QinetiQ have decided to play an active role in the OSS/J Initiative and now sit on the Initiative's Advisory Council.
What's more, the TeleManagement Forum has acknowledged OSS/J APIs and the first - and to date, the only - implementation of NGOSS. Indeed, OSS/J technologies were an integral part of four major TMF catalyst projects on display at this year's TeleManagement World conference held in Nice, France, May 16-19:
* The OpenOSS Catalyst -- a testbed of freely available OpenOSS Initiative open source components.
* Business Agility Implementing the SID Catalyst --Â© Use of the SID model to simplify and accelerate integration-related development in a service activation scenario defined by France Telecom.
* Business Activity Monitoring (BAM) Catalyst - providing BAM over the OSS/J Service Activation, Trouble Ticket and Quality of Service APIs.
* NGOSS/MDA: Realising NGOSS as a Model-Driven Approach - using a model-driven problem-solving methodology and leveraging state-of-the-art standards and tools
Agilent's Mottishaw has been an active participant in the OpenOSS Catalyst Project which, he says, "We wanted to enable participants from across the industry to collaborate on the implementation issues that we face in delivering OSS standards based solutions. Proprietary software makes this difficult because nobody wants to expose the implementation issues they face. Open source software solves this because the source code is available to everyone. OpenOSS is not attempting to compete with proprietary solutions, but will help the industry address some of the integration problems that limit the value of current products." In fact, OSS/J APIs were a critical component of the project.
"OSS/J's QoS and Trouble Ticket interfaces play a key role," Mottishaw says. "We successfully used OSS/J to integrate open source OSS software and COTS OSS products in a complementary way to create an initial sandbox for further development projects."
The OpenOSS Catalyst is especially significant because it is backed by a consortium of industry leaders -- BT, COLT Telecom Group, Covad, NTT, and QinetiQ -- as well as Britain's University of Southampton and Hungary's Budapest University of Technology and Economics. The purpose of the OpenOSS Project is to:
* Embrace the TMF NGOSS approach to develop a 'sandbox' of open source software
* Make software and associated documentation freely available to the industry
* Provide feedback on implementation experience into OSS standards activities
* Provide a vehicle for research to engage with realistic service provider problems
(More information on the OpenOSS initiative which delivered the TMF Catalyst can be found at http://www.openossinitiative.org/.)
Craig Gallen, who leads the University of Southampton's OpenOSS Project agrees, adding that OSS/J and NGOSS are reshaping the market.
"OSS/J has grounded and realised the work of the TeleManagement Forum and I think that's very important," he says. "The Initiative has proven that low-cost components from the enterprise world can be easily integrated into the OSS/BSS environment. This will lower the barrier to entry for non-incumbent component vendors -- effectively commoditising parts of the OSS market while allowing incumbent ISVs to concentrate on delivering high-value solutions. And, finally, OSS/J provides a framework for sharing service provider investments in software development." n
William F. Wilbert has written for technology publications for over 15 years. He is currently Communications Manager for the OSS through Java Initiative, and can be contacted via e-mail: Wilbertz1@comcast.net