Frame relay gets a jolt

Meeting needs is crucial to market staying power and that is exactly what the Frame Relay Forum has in mind for frame relay networks. A recent implementation agreement ratified by the forum gives frame relay the ability to support Sonet/synchronous digital hierarchy, thus enabling support of higher capacities.

The Physical Layer Implementation Agreement, or FRF.14, raises the bar from speed limits of T-3 (45 Mb/s) or E-3 (34 Mb/s) to OC-3 (155 Mb/s) and OC-12 (622 Mb/s). This agreement came after users, service providers and vendors pushed for higher access speeds and greater remote and dial-access scalability. One concern of the forum was due to frame relay's limited capacity and limited scalability; it would lose ground to alternate technologies like asynchronous transfer mode (ATM).

"Until FRF.14, frame relay had been running basically at DS-3 speeds, which limited network access to DS-3," said Melanie Hansen, vice president of international development for the Frame Relay Forum and executive manager in virtual data services marketing for MCI WorldCom. With backbones in the carrier's networks getting so large, even DS-3 is not acceptable, Hansen said. "Carriers have had to deploy ATM in their backbones simply to achieve scalability for users."

Because of the increased capacity and scalability allowed by FRF.14, that trend of resorting to alternate technologies may change.

"A lot of carriers may have already put ATM in their backbones and that probably means it will stay there, but for carriers that haven't upgraded their networks-they will be able to stay with a pure frame relay solution that can scale higher," Hansen said. Bringing a whole new technology and management system in just to get more bandwidth doesn't make sense when frame relay can achieve OC-3 and OC-12, Hansen added. "By keeping it all frame relay then they will be able to minimize their investment and still get scalability," she said.

Frame relay has come a long way from its humble beginnings as a LAN-to-LAN interconnect service operating at speeds of up to T-1," said Christine Heckart, vice president of telecom management for TeleChoice.

"We have seen frame relay's scalability grow along with its popularity to the T-3 level, and this development will ensure a continuing long-life for frame relay networks."

But that notion of frame relay's lack of capacity is really a fallacy, said Larry Greenstein, vice president of the forum. "There's been a misunderstanding from the very beginning that frame relay is limited to speeds of 2 Mb/s. It can go as fast as the [equipment] technology can push it," said Greenstein, who is also director of product management for Nuera Communications.

Hansen says that while the increased capacity is important, the interoperability induced by the agreement is just as crucial. The previous frame relay agreements included references to the physical layer, but a complete guideline did not exist, so FRF.14 cleared the confusion by addressing the interfaces (see table).

Maintaining interoperability was easy to accomplish with this agreement, Greenstein said. "This one came in early in its evolution so there wasn't much contention," he added. The forum runs into more difficulty when a technology or product is already evolved and has been deployed or when there are multiple implementations, he said.

"We often see that vendors are doing parallel development [as in this case], so there will not be an enormous time lag before the implementation starts to happen," Hansen said. Implementation will start later this year, she added.

But even after the capacity and interoperability enhancements, is there enough market share left to grow frame relay? Both Hansen and Greenstein agree that the rate of growth for frame relay networks is leveling off because the market is saturated, but it still represents a significant services revenue stream.

"It's hard to grow 100% every year, although user uptake in the service area has been very robust," Hansen said.