Gateway top voip

No matter how a vendor sees SBC (session border control), one thing is certain: The need that emerged in 2003 has become a demand from every major carrier running voice traffic over IP networks. Session controllers enable voice traffic by bridging between various IP networks--primarily H.323 and SIP (Session Initiation Protocol)-based--and by interworking with different softswitch and router technologies while overcoming the barriers installed in networks to protect data transport.

The SBC vendor lineup can be broken down into three basic types:

* Standalone players. Suppliers, which recognized early that transporting voice over IP networks was a tricky business and developed technology to overcome bridge firewalls and NAT (network address translation) barriers. They hope to sell or license their products to carriers or larger vendors to be bundled into softswitches, routers or other voice system components.

* Traditional voice technology vendors. Through in-house development or acquisition, these firms plan to incorporate SBC capability into their over-all technology.

* SBC technology companies. Vendors that hope to develop SBC technology to attract the highest bidder and become part of a larger entity.

"Everybody says there is a clear need for session controllers, and there is no doubt they will be deploying them," said Micaela Giuhat, executive vice president of product management for Netrake. "The session controller will look into the messages, extract the information it needs from the payload regarding the calling party, dynamically open a firewall, let the call go through and do all the address translation that you need."

The best way to do this is to build a separate, interoperable product, said Giuhat, because "you always have better performance."

Companies such as Singapore-based MediaRing and NexTone Communications have done just that. Dan Dearing, marketing vice president at NexTone doesn't argue the point about standalone products having better performance, but he thinks session control is more than an edge technology.

"Session control is a multi-dimensional problem," said Dearing. "It includes SBC at the edge of the network, but it is actually a much richer set of solutions that includes how you route traffic as it enters and leaves the network and how you manage it."

NexTone partners with SnowShore, which has a programmable media-processing platform that fits "very nicely into the demarcation between the signaling guys like NexTone and our media platform," said Joel Hughes, SnowShore's CEO.

SnowShore adheres to a softswitchcentric approach. "We look at Nortel. Siemens. Alcatel, people who are selling softswitches and gateways and believe they will simply add a new gateway-like platform--in our case a media processor--to their product suite, but they'll continue to leverage their core softswitch and extend that to address session control issues," Hughes said.

The core versus edge argument doesn't get much traction with Jim Greenway, vice president of marketing at Kagoor.

"To me, core and edge is more location in the network. The primary functionality that we're focused on is carrier-to-carrier and carrier-to-customer," he said. "We think that true SBCs are purpose-built. As more time goes on, more features get heaped on these things; only the ones that are purpose-built for this job are going to be survivors and winners."

The problem with purpose-built products is that they add components and cost to the network and "the last thing any service provider wants is another product in its network," said Jim Hourihan, vice president of marketing and product management for Acme Packet. "The reality is they need this functionality."

While a modified softswitch or router could provide SBC, a standalone product does it better, according to Hourihan. This is compounded by the nascent nature of the space where smaller, entrepreneurial companies with less infrastructure can respond more quickly to shifting customer demands. Big companies will eventually take over SBC responsibility, said Nathan Franzmeier, CEO of Emergent Networks.

"I don't believe that a session controller as a standalone entity is a long-lived product," he said, even though that's what Emergent builds.

Emergent also builds softswitches, so it's an easy view to take. It's not so easy for others to dismiss standalone session border controllers.

"It's a new box that is not under the control of another softswitch," said Michael Rubin, director of product management at Sonus Networks. "It interacts with a softswitch ... but it is not under the control of the softswitch; it is not an incremental add. It's a completely new box."

Nortel Networks, being a big player in the softswitch space, takes a more integrated view. SBC is really "distributed border control," said John Egli, director of VoIP marketing for Nortel Networks. "We don't actually have a separate box. We have that function distributed across other network devices."

Dan Freedman, CEO and founder of Jasomi, believes that big companies like Nortel and others must acquire SBC expertise, since it's too expensive and time consuming to develop. He blithely lists the reasons why his company should be at the top of the shopping list.

"We can be bought for somewhere in the $30-million to $100-million range, whereas none of the other companies in this space can be this year," he said. If a sale's to be made, it should happen now, he added.

"Over the course of two or three years, there will be less and less reason for people to want to solve the problems that session border controllers solve today on a new piece of hardware," Freedman said. "Instead, they will go with their softswitch vendor or their fire-wall vendor to get this problem solved."

Jasomi, he suggested, is unique. It's making money and has little debt. Other start-up companies in the space have high debt and no profit.

"We are focused on enabling consolidation of the SBC market by presenting the only company in the space that is absorbable this year," he said.

That's a solid strategy if the softswitch maker wants the component and the carrier wants a unified offering. For much of the market, though, a SBC may be like a CD player in a home entertainment system: a specific component with a specific purpose that it can fill better alone than as part of the whole system.

"There are still a whole lot of questions, and I don't think anybody has the answers," said Christine Hartman, vice president of voice-over-packet markets for the Probe Group. "I think a lot of the big guys are looking into it, and they're going to change the architecture."

The early architecture seems to be influenced by the need for product development speed and specialization.

"SBC vendors are solving a small piece of the entire puzzle," said Alan Bugos, vice president of engineering for VoIP carrier iBasis. "They're very quick to move and react to any kind of feature changes we request."

And that, for now, may be what this market needs. Beyond that, the subject's still open for discussion.

"I think the whole category is in flux," said Hartman. "It starts to shake out this year, but it's going to take some time."

Jim Barthold is a contributing writer for Telecommunications[R] magazine (jbarthold@comcast.net).