Once the domain of the deeply nerdy, operations support systems are moving out of the shadows.
Once the domain of the deeply nerdy, operations support systems are moving out of the shadows. Analysts suggest that even as spending on network equipment slows, the OSS industry will continue to grow 5 percent to 15 percent this year.
Among the drivers of this growth: the addition of wireless, packet-switched, all-optical, broadband and other services, spiced up with a rate of change that could hardly have been dreamed of in 1984, when Bellcore was established to keep all the regional Bells singing from the same operating hymnal.
"Networks of complexity really begin to put a strain on the OSS foundations" of traditional telecommunications companies, says Ron Ponder, who heads the telecom practice at Cap Gemini Ernst & Young.
Ponders colleague, Greg Douglass, who heads Cap Geminis OSS practice in Dallas, sums up the problem more bluntly: "Their business is just cratering under the weight of these new services."
To speed return on their hefty network investments, service providers need to be able to add new services quickly -- without spending a ton of money or alienating the customer in the process.
"They need to get to higher-value services in order to differentiate themselves," says Martin Steinmann, vice president of marketing with Syndesis, a fulfillment systems vendor based in Toronto. "Services are very complex to deliver in a cost-effective manner. Moving up the value chain is a huge challenge for service providers."
OSS software that works well is an essential part of the equation. It can streamline order entry, provisioning, quality control, inventory and network management, and other processes.
Interoperability testing to prove incumbent carriers readiness to compete in the local loop has put an unusual spotlight on one aspect of the OSS world: connecting traditional carriers to competitors. But others will take center stage as all-optical networking grows in importance.
"I spent many years as a [chief information officer]" Ponder says. "Oftentimes, those CIOs are guilty of spending a lot of their money on sophisticated billing systems or their networks, and overlooking the OSS, or taking it for granted. It just sort of existed in the back room."
A New Back Office
Perhaps the biggest change agent in the operational support area is the industrys gradual migration onto Internet Protocol (IP) and all-optical networks.
According to those in the industry, traditional OSS approaches fail because theyre based on a point-to-point, circuit-switched network, rather than a mesh architecture that collects and distributes packets. To provide the data, network management and service levels that companies expect, these systems must gather information from many different directions and make it look seamless to the end user.
What Communications Industry Researchers analyst Lawrence Gasman calls "the Nimble Network" will require much closer attention to software.
"The whole issue of OSS and network management really needs to be reexamined in the light of optics," he says.
Legacy OSS were engineered to run voice over copper, not packets over fiber, in an era where cost control was less important than building a network for the ages, Gasman says. "As the transport layer changes to the new designs, the old OSS designs become less and less relevant."
New designs will focus on making the most of assets: fiber inventory, bandwidth and guaranteed service levels. In a speech at the 2001 Optical Fiber Communication Conference last March, Gasman predicted that optical performance management software will become the second-largest segment of the OSS market, behind provisioning products for the all-optical network.
Provisioning products will remain the largest segment, growing from $500 million in sales in 2001 to about $825 million in 2004, according to research by CIR. But performance management will see robust growth as service providers use the tools to create and monitor service level agreements and check the health of their networks.
Other software categories, such as fault management, network planning, inventory and billing, will grow as well, Gasman says. "Network nimbleness is almost synonymous with software."
Much of the blame for the apparent disarray in OSS has been laid at the feet of legacy systems. Many OSS applications shipped with hardware, leaving integrators to patch a dozen apps together along with a dozen network components. And hardware companies often arent the best developers of software applications.
"Thats really what slowed down Telcordia [Technologies]" Douglass says. "It got too big and unmanageable, and it couldnt be flexible."
Telcordia -- formerly Bellcore -- was created as part of the mid-80s Bell breakup to ensure common network standards across the newly independent regional Bells. These days, at Telcordia headquarters in suburban New Jersey, engineers are laboring to prove that old dogs can learn new tricks.
Telcordia, which was purchased by Science Applications International Corp. in 1997, inherited Bellcores legacy: big, profitable, first-tier customers based on big, old, circuit-switched networks. "Cherished systems," Chief Architect Mark Effinger calls them. "Legacy" carries with it a certain negative connotation."
The positives: The regional Bells operate the worlds most efficient telecommunications systems, theyre hungry to add next-generation ser-vices and they have plenty of cash flow and customers.
Effinger says Telcordias not suffering from the demise of many new-generation carriers because they were never Telcordias prospects in the first place. "They werent focused on operational efficiency. They were just dressing up for Wall Street."
The risk, Effinger says, remains the inadvertent -- or unavoidable -- creation of "silos" of network management and support.
"You wont hear Telcordia announce our all-optical OSS," Effinger predicts. Instead, the company will continue its current strategy of adding managed IP solutions on top of its existing capabilities, recognizing that smart networks can handle some back office tasks themselves instead of delegating them to an OSS. Two such products are scheduled for release this year.
"You cannot possibly afford to patchwork," he says. "It just breaks down."
Make Money Now
Wall streets message to telecommunications companies is coming through loud and clear: Take the network that youve been spending money on for the last five years and start making money on it -- now. Companies are scrambling to run more traffic on high-speed networks at lower cost, add more high-margin services before competitors and take advantage of as much automation as they can.
If incumbent carriers pass their competitive tests and start offering long-distance under Rule 271 of the 1996 Telecommunications Act, OSS applications will get the credit.
One of the more aggressive carriers in its pursuit of lost long-distance business, Qwest Communications International has been using OSS from Syndesis to provision digital subscriber line circuits automatically. Qwest inherited the DSL business when it bought U S West, which led the industry in rolling out DSL, only to see service complaints grow as fast as demand for the high-speed service.
Errors abounded in the network databases. More than of 50 percent of the DSL orders coming in to U S West required some degree of manual processing. Workers would provision circuits based on outdated inventory data. Technicians would discover the circuits didnt really exist only after theyd rolled a truck. U S West had plenty of company; NorthPoint Communications fallout rate -- problems that need to be dealt with by a human -- was 60 percent before it collapsed, Syndesis Steinmann says.
After the Syndesis NetProvision implementation went live, average provisioning time went from 40 minutes per circuit to 30 seconds. The fallout rate fell to 2 percent or 3 percent. Real-time troubleshooting enabled technicians to check -- and often fix -- problems while the customer was still on the telephone.
Syndesis expects to prosper by focusing on the needs of service providers in the metro market that are looking for the holy trinity of benefits: service innovation, cost savings and quick return on investment. For companies investing in these kinds of systems, Steinmann points out, "the payback is literally in weeks."
Telcordia has also approached the problem, Effinger says. Its Holy Grail is to build software that enables its clients to deploy any service over any network -- cable TV over fiber, video over DSL, phone over cable -- and make money at it.
"These services are radically changing, constantly changing," he says. "The network is constantly changing. What hasnt really changed is the focus on profitability."
Dealing with the various integration issues facing telecom providers remains a nightmare. Adding new services is just one integration problem; service providers also must cope with equipment changes, consolidation -- in their industry, and in those of their customers and vendors -- and new technological underpinnings.
"In the IP world, you have to assemble data from a bunch of sources," explains Jim Alsman, OSS program leader at Frost & Sullivan. "Theres still these legacy silo systems. Thats still a large problem and until its solved, the markets not going to go anywhere."
Some of the strategies to navigate integration include buying the kind of one-stop-shop product suite that vendors increasingly are striving for, such as Syndesis NetProvision, which includes modules for service activation, order management, billing, fulfillment and other functions.
"We believe its still a fairly scattered market," Steinmann says. "That makes integration impossible."
Another approach involves retaining a consultant like Accenture or Cap Gemini which can provide a set of pretested best-of-breed products, ready to go. Cap Geminis Douglass describes his companys structured approach as including areas often neglected by system integrators, such as business process analysis, data conversion and training on the new system.
A typical integration project, Douglass says, can cost $10 million in new hardware, software and services; take six months to implement; and leave organizational chaos and unhappy customers in its wake. Cap Geminis approach is to "deliver incremental functionality," getting pieces of the system up and running in stages, starting with network management. Theyre also pushing software vendors to deliver applications that are more flexible and better suited to support new telecom products.
Siebel Systems is one enterprise software company thats moving into the telecom arena with product lines specifically addressing the needs of carriers. As Siebels largest vertical market, with $320 million in licenses last year, telecom rated its own "multichannel eBusiness application" -- Siebel eCommunications.
The communications suite adds billing, fraud protection, pricing and other telecom-specific modules to the core sales and customer service functions. According to Dan Ford, general manager at Siebel, the goal is to broaden carriers network-centric view of the world "How many more bits can I pack on the line before I have to add another switch to encompass customer relationships.
"The relationship with the customer is really a profitability issue," Ford says. "Theres a lot of customer information weve been getting on the front end that could be really valuable to network planners."
Software vendors -- and, to a lesser extent, hardware vendors -- are also doing their part to create integrated solutions through mergers and acquisitions. In 2000, Amdocs paid $1 billion for Solect Technology Group and Sema Group ponied up nearly $4.7 billion for LHS Group.
Earlier this year, ADC Telecommunications paid $186 million for CommTech, while CrossKeys Systems network and service assurance apps were worth more than $35 million to Orchestream. The new Orchestream is focusing on virtual private networks and other IP offerings, building multivendor products that are independent of the network below them, for easier provisioning and service, says Steve Adams, vice president in charge of the Resolv product line.