TMCnet
ITEXPO begins in:   New Coverage :  Asterisk  |  Fax Software  |  SIP Phones  |  Small Cells
 
 


December 2006, Volume 9/ Number 12

Return to IT Table of Contents


The Year in Review...The Year Ahead

By Richard �Zippy� Grigonis

 


2006 was a pivotal year in the history of IP Communications. Practically all greenfield installations are now IP-enabled, and many existing circuit-switched phone systems are being modified or replaced with systems that can deal with both the PSTN and IP networks. 2007 will give us more of the same, along with new technological advances in wireline and wireless communications, and applications having myriad new features.

With the new IP technology came a new crop of buzzwords and terms. Web 2.0 was joined by the somewhat more nebulous Web 3.0. As one online commentator wrote, �Web 3.0 is Web 2.0+... more of what we have now, but faster and more ubiquitous.� It was also linked with another term, the Semantic Web, according to Dave Linthicum. (Wikipedia says that, �The Semantic Web is a project that intends to create a universal medium for information exchange by putting documents with computer-processable meaning [semantics] on the World Wide Web.� It uses XML, XML Schema, RDF Schema and OWL.) And SOA, or Service-Oriented Architecture, was on everyone�s lips (though not everyone knew why).

Despite this multiplicity of acronyms and interesting terms, we saw more of the same in 2006: More Skype users, more open source (Asterisk-based) phone systems, more sophisticated wireless devices, more WiFi (News - Alert) and dual band phones, more broadband, and more SIP-based systems, thus cementing the position of this important call control/signaling protocol.

The public continues to clamor for broadband, despite the fact that many U.S. carriers have been slow to respond by changing their infrastructure, at least by European and Asian standards. Call/contact centers began to take IP Communications seriously, even though VoIP adoption rates in such centers had just barely entered double digits toward the end of 2006.






Business-wise, we saw one IP IPO that fizzled big time (that of Vonage) and another that did extremely well but stayed under the public�s radar (that of Acme Packet (News - Alert)). One bellwether company in this industry, Cisco Systems, revealed terrific revenues � see Rich Tehrani�s Publisher�s Outlook on page 8 of this issue.

Video mania took hold in 2006. The public eagerly awaits IPTV (News - Alert), the classy, interactive PayTV version of IP video. In the meantime, however, it seems that everybody from well-known motion picture and network TV companies to teenagers and an assortment of lunatics have generated zillions of Internet videos of varying quality. Blogs have been supplanted by video blogs or vlogs. Video is even going mobile. Indeed, cell phones are becoming personal mobile multimedia and information stations, as established media companies try to repackage existing content for new delivery environments and, therefore, new distribution channels. For example, Hearst Magazines struck a deal with Volantis, a leading supplier of Intelligent Content Adaptation� solutions, to launch mobile device-savvy versions of Hearst Magazines� Cosmopolitan, Seventeen, and CosmoGIRL!. The magazines will be mobilized utilizing Volantis� Mobile Content Framework� and will soon be available on both the Cingular (News - Alert) Wireless and Sprint-Nextel networks.

In related news, America began to fall in love with mobile broadband and the AirCard, though existing over-air connections may seem anemic when compared to upcoming WiMAX deployments in 2007.

IMS, the IP Multimedia Subsystem (News - Alert), destined to yield a common service architecture for both wireless and wireline, and promising the ability to easily create and deploy a multitude of new services, stayed mostly in the interoperability laboratories. More extensive trials and deployments will appear in 2007. As they do so and the world�s communications infrastructure is renewed, you�ll see heavy duty computing platforms appearing in the network based on AdvancedTCA (News - Alert) and MicroTCA form factors and running switch fabrics such as 10 gigabit Ethernet, PCI Express, and Rapid I/O. You�ll even see some workhorse media servers in the form of older CompactPCI (News - Alert) and PCI-bus technology.

Of course, the above verbiage is just one person�s observations. Yours Truly decided to take his inquiry to the next level by talking to a number of well-known and respected companies in our industry, gathering together and distilling their opinions on what really happened in 2006 and how IP Communications will bode in 2007.

8x8 (News - Alert)

Despite the pall over the industry cast by the Vonage IPO, many service providers have been doing rather well.

Bryan Martin, Chairman and CEO, 8x8 Inc. (www.8x8.com) (the folks who bring you the sophisticated yet inexpensive Packet8 services) says, �During 2006, while we continued to see the adoption of Internet telephony motivated by price, we also saw consumers and businesses looking to IP Communications for new services and applications that were never available to them before through legacy network offerings. The incorporation of video and business service applications as part of the IP voice experience was prevalent throughout the year as customers began to value Internet telephony for reasons beyond the phone bill at the end of each month. 2007 holds a renewed promise that the industry will use IP Communications and the Internet as a means to develop away from the hard, closed walls of the legacy voice networks.�

Martin adds, �It is my sincere hope that the regulatory community and lawmakers will also come to realize that we cannot just find ways to shoehorn past regulations and policies onto these new applications, as was done throughout most of 2006 with E911, universal service, and intercarrier compensation reform, but to really begin to think in new ways and use the canvas of the Internet to draw better, improved solutions for the communication networks, services and applications that will become the prevalent means of communication for our society throughout the 21st century.�

Acme Packet

Unlike Vonage (News - Alert), Acme Packet (www.acmepacket.com) had an IPO success story in 2006. It closed its initial public offering of 11 million shares at $9.50 per share, raising about $110 million. Yours Truly remembers when Acme Packet�s affable CEO, Andrew D. Ory, left Boston Technology and formed Priority Call Management in 1991, a leading supplier of enhanced calling and messaging solutions, which later became a division of Atos Origin. Ory�s move to Acme Packet helped bring increased success to the company.

Acme Packet�s VP of Marketing and Product Management, Seamus Hourihan, says: �When looking back at 2006 and looking ahead at 2007, I see developments in five areas. First, there�s Architecture. I think we�re moving from IMS as dogma to what I call �IMS pragmatism�. I like to play with acronyms. IMS means a lot of things to many different people: To service providers from a business perspective, IMS stands for �Incoming Money Soon�. From a vendor perspective, it means, �Install More Sh*#�. If you�re on the operations side, trying to make it work, it means, �Inflict More Suffering�. And in the background across all of this, it means that the �IETF [Socialistic] Mindset Sucks�. As you can tell, I�m not politically correct, I�m very irreverent.�

�On the pragmatism side,� says Hourihan, �as my recent article on IMS security in IMS Magazine mentioned, IMS is basically missing security. People need to be pragmatic about how they fill those gaps. We�re also seeing pragmatism in terms of some of the other standards organizations taking the 3GPP architecture, extending it and modifying it as they see fit. That would include ETSI TISPAN, the PacketCable group and MSF [Multiservice Switching Forum].�

�Again, there are many things missing in IMS,� says Hourihan. �I�m working on an article basically entitled �Does IMS Work?�, and that set of concerns relates to IMS scalability when it�s deployed to serve many subscribers. It also includes things like call flows and the complexity surrounding them; interactions with databases and how often they occur and if there are there better ways to do them � perhaps not on a per-call basis, but more in terms of pre-allocated permissions and various network elements.�

�We�re moving from this dogma of �IMS is God� to the realization that it�s not perfect,� says Hourihan. �We�ve got to be pragmatic about how we deploy IMS, not only in different types of networks � cable versus wireline versus wireless � but also pragmatic from the perspective of each individual service provider in terms of where they are today and where they�re going to end up.�

�Second, there�s Services and Applications,� says Hourihan. �From an IP Communications perspective, by and large, we�re still stuck in �basic black phone service�. In fact, as I called you today, I got angry because of the number of digits I had to enter to connect to this conference call. On the other hand, in a Unified Communications IP world, I could just click on something and it would get me into the conference. And it would be a video conference.�

�We�re already seeing among our customer base people who aren�t waiting for the deployment of full-blown IMS architectures to do some interesting stuff,� says Hourihan. �For example, we have a number of providers, such as Telefonica, Mobilis, Vodafone Spain, and Chunghwa Telecom (News - Alert), that are offering fixed mobile convergence [FMC] services today, in the form of �video telephony� if you will, across wireless and wireline environments. We also have many customers outside the U.S. offering a collection of rich SIP-based services, such as interactive video, today � Telecom Italia (News - Alert) being one.�

�What�s interesting to me is that all of this innovation moving toward Jeff Pulver�s �purple minutes� concept, is that none of this activity really is happening in the U.S.,� says Hourihan. �It�s all outside. I think there are some cultural and demographic factors indicating where these services are really taking form. I don�t mean to be politically incorrect in this next comment, but if you think about the people from Spain and Italy, well, they love to talk, they love to see people. Some Europeans just have a more natural inclination to embrace some of these new features offered by IP Communications, and that�s one of the major drivers.�

�The U.S. is, quite frankly, in the dark ages,� says Hourihan. �We�ll eventually start to turn the corner, but it�ll be slow, propelled a great deal by the same youth movement that drives YouTube and MySpace. It�s a matter of how providers market themselves.�

�Third, is Net Neutrality,� says Hourihan. �I think this is a non-issue. We already see increasingly the delivery of tiered network services and the emergence of federations. In some cases, the initial goals around federating will be to save money. For example, cable operators moving to federate and exchange traffic via IP to save money by keeping the traffic off of the PSTN. Some of this begins with peering exchanges, such as Stealth Communications� VPF [Voice Peering Fabric] in New York, a distributed exchange allowing members to establish peer-to-peer connections for VoIP traffic, or XConnect (News - Alert)�s peering service for over 400 VoIP service providers. The motivation to save money will change over time, and instead of just connecting everybody over the public Internet, providers will supply trusted secure services and ensure high quality end-to-end service as well. For example, how do I know when I make an IP phone call that the call actually gets from me to, say, the customer service agent at the Bank of America? And to do it over a quality connection so that when I give them my social security or account number over the phone, I don�t have to repeat it ten times.�

�Fourth, are Regulatory issues, the regulatory requirements on IP Communications,� says Hourihan. �Looking back over the last few years at FCC (News - Alert) regulations in the U.S., we are moving from what I call a reactive �God forbid� regulatory environment to something that�s more proactive, but still driven by �God forbid� scenarios. This environment started when Janet Jackson had her wardrobe malfunction at the Super Bowl. Basically the outcry was, �God forbid that my children see that again on TV�. And so we found ourselves with a new set of decency laws applied to broadcast television.�

�The next �God forbid� situation involved Vonage,� says Hourihan, �when subscribers died in Texas and Florida because they couldn�t reach their respective 911 services. In this regard, the FCC has moved from the reactive mode to the proactive mode, demanding customer access to E911 service. Also, early in 2006 they announced that all VoIP service providers must provide lawful intercept capabilities. Why? Because they�re again being proactive. �God forbid� that Osama Bin Laden uses Skype (News - Alert) to set up his next terrorist attack in the U.S.�

�There�s something else on the horizon with respect to regulatory, and that�s the issue of taxation, specifically state taxation,� says Hourihan. �PSTN-derived tax revenues provide universal service and, more importantly, support state funds that finance such things as schools. As tax revenue from the PSTN declines, �God forbid� we soon don�t have money to educate our children. Bottom line: I don�t view that IP Communications will continue to be exempt from many aspects of regulation.�

�Fifth, and finally, there�s Competition; that is, the competitive environment both from a vendor perspective as well as a service provider perspective,� says Hourihan. �I think the trend will be increasing consolidation in both areas. From the vendor perspective we�re already seeing the beginnings of this with jokes about �Alca-cent� and �Nokia (News - Alert)-mens�. In the IP Communications area, we have a very fragmented vendor environment. Acme Packet partners with 13 telecom equipment makers on a worldwide basis. None of them are dominant from my perspective, and we�re starting to see these 13 collapse to 11. More of this will occur over time. Our session border controller marketplace today is populated by about 20 vendors. Clearly, a mature market won�t have that many players.�

�On the service provider front I think things are more fragmented,� says Hourihan. �In the U.S. you see the incumbent telcos such as the AT&Ts and Verizons, then there�s the cable companies, and ultimately some type of third provider with a technology such as EV-DO or WiMAX, be it a Clearwire or a Sprint (News - Alert), will emerge as being important. The key for making IP Communications work will be the facilities. Even in a world of fiber-to-the-home or WiMAX-type bandwidth, IP Communications will increasingly support voice and video interactivity. That requires new equipment in the access network.�

�There will still be Internet-based services,� says Hourihan. �But again, if you�re looking for trust and quality of service [QoS], you�ll be looking to facilities-based providers that allow anonymous subscribers. At Yahoo or Google (News - Alert) �I can be a dog on the Internet�, to quote that famous New Yorker cartoon, along with, �Nobody knows who I am or what I am�. Some providers will differentiate themselves in that they will know their subscribers, at least in terms of who they are, where they are and what their income level is.�

Aculab (News - Alert)

Aculab (www.aculab.com), a well-known maker of hardware and software building blocks for telecom, builds devices that must stand at the crossroads of the PSTN and IP Communications. Aculab�s Sales and Marketing Director, Chris Gravett, says, �If anything, 2006 was the year when we realized that the IP market had moved beyond the hype. In the second week of January 2006, we launched our first true pure IP product, the Prosody X. That meant that we were able to start responding to opportunities that were beginning to appear back in 2005. We found that we are good at dealing with real commercial opportunities, and they are from major companies that have made the decision to release new platforms and solutions that are IP-friendly or have the IP capability built in. So we�ve seen companies across the world willing to invest in the technology and start planning their own next-generation releases. 2006 has very much been a year of intensive development and the release onto the market of solutions by our customers.�

�It�s almost unheard of now for a new opportunity to be only TDM in nature,� says Gravett. �At a minimum, it has to be IP-ready, which is where Prosody X comes into its own. The other thing pertinent to IP Communications during 2006 is that it has become apparent that voice alone is not a sufficient �sell� for IP Communications. We�ve come under a lot of pressure and demand to deliver a broad range of codecs, particularly addressing video and mobility opportunities. Security has also become an issue and, therefore, an opportunity from our point of view. I�m talking here about things like Secure RTP.�

�So 2006 has really seen the promise of the previous five or six years actually come to fruition,� says Gravett.

�Something that�s happened a bit more slowly than we anticipated has been the move to host media processing and the corresponding move away from DSP-based media resources,� says Gravett. �At the end of 2006, the conclusion we reach is that, not surprisingly, there is still going to be a future for DSP-based resources. However, over the next couple of years � and particularly as we move to a pure IP environment � the SMB area of our traditional market will much more be accommodated by host media processing. The demand for wideband, video codecs, and so on will ensure that there is going to be a need for higher-powered DSPs and a lot more of them. In an IP-only environment we see the need for probably AMC [AdvancedTCA Mezzanine Card]-based DSP farms, both for the large enterprise and for the telco market. We�re studying that upcoming market at the moment.�

�However, at present we have not signed off on any development projects involving AdvancedTCA equipment,� says Gravett. �We�re very much aware that we probably should have something going in this area. I suspect that before the end of 2006, we will have embarked on one or more ATCA projects. One thing that�s been holding us back is that we would like to be utilizing higher horsepower DSPs, which are only just now becoming available, even for development purposes. If we embark on an AMC development, then it�s not a good idea to use today�s DSP technology. But I envisage that we will have development projects underway by end of 2006 and the beginning of 2007. We�re certainly heavily involved in the technical and market research to bring that about.�

�One question we have to answer is whether the density that we�ll likely to be able to achieve using DSPs will exceed what can be achieved with host media processing,� says Gravett. �The question marks surrounding host media processing concern not just the density in terms of processor power, but the footprint size and heat considerations when one builds top-end systems. That�s an area where we�ll spend a lot of time comparing the pros and cons before we embark on what will be a big investment for Aculab, the move into the ATCA arena. My personal belief is that our entry into the ATCA market is inevitable. It�ll be sooner than later.�

BEA Systems (News - Alert)

Mike McHugh is VP and General Manager at BEA Systems (www.bea.com). He�s responsible for driving the WebLogic communications platform for enterprise applications and SOA (Service Oriented Architecture) � which includes BEA WebLogic Server� 9.2, BEA WebLogic Portal� 9.2, BEA WebLogic� Integration 9.2, BEA Workshop� for WebLogic 9.2, and BEA JRockit� 5.0.

�VoIP isn�t �the thing� any more, simply because VoIP has pretty much become mainstream,� says McHugh. �This year, it wasn�t so much asking �Is this stuff going to happen?�, it was more like, �How do we leverage it now that it�s here? What are all the new services?� You go to expos and there�s hype around new services. The good news is that this year we�re beginning to see those services.�

�Here in San Francisco,� says McHugh, �Comcast (News - Alert) is a local cable provider, and they�re running radio ads about triple play. �Sign up here for your $69.95 triple play� and they actually use the term �triple play�. So even the terminology is becoming mainstream.�

�Fixed/Mobile Convergence [FMC] was good talk but not �real� until all the players came on board,� says McHugh. �So you�ve got Sprint/Nextel, AT&T, BellSouth (News - Alert), Cingular, and some of the new entrants that are going to make triple or quad play real. The whole IP video field also made a big impression this year, such as Google acquiring YouTube. It all underscores the arrival of IP-based capabilities and the fact that VoIP is old news now. This year we saw the emergence of the push to identify what to do with IP Communications now that we have it. What kind of increasing richness can we offer in the more interesting applications as they become mainstream?�

McHugh elaborates: �This year BEA started �beating the drum� about how you go about leveraging IP and how enterprises and operators are going to offer services and what�s that going to look like. We realized we had some ideas about what it takes to build platforms that run these things. As IP comes online, and the Internet and many forms of communication merge, we see the emergence of some very familiar paradigms of what it takes to build these. At least it�s familiar to us, having lived through the Web transformation over the prior decade. Some of the data and services management paradigms emerging we think are directly applicable to IP. That includes concepts as IT-sounding as SOA � the service oriented-architecture � that enablement of services reuse and services infrastructure use. It�s something we very much believe in and we started beating the drum for that in 2006 too. Next we�ve got to deliver on all of this and show with �proof points� how it works.�

�As for 2007, there are two elements to �what�s next�,� says McHugh. �There�s the infrastructure for building it, and then there are the services that might use that infrastructure, both for consumers and the enterprise. You see common applications at the shows, such as gaming-type applications, and in the business space they tend to be more person-to-person collaboration-type communications enablement or extensions to traditional collaboration business-type applications. We think that the increasing pace of arrival of these services and the demand for services drives the value, but, in our view, it will take some rational infrastructure to pull this off. If you�re sitting at home surfing or writing something, you�re not thinking of the network underlying that IP system. It doesn�t matter if the transport is cable or DSL or whatever. What matters to you is that the service is available and what you can do with it. �

�We�ll see new infrastructure and cost models that enable quick time-to-market, easy development and deployment,� says McHugh. �It�s the notion of throwing a lot of services at the wall and seeing what sticks. The cost of failure can�t be very high, because if the service doesn�t �stick� or find favor with users, then you�ve got to throw it away and move on to other services or other combinations of services, until you find something that works for you.�

�We believe that communications will become highly personalized,� says McHugh. �When I talk to you, I don�t want to look up your phone number. I want to talk to you. That idea could result in an application where the system calls you on your IP phone or multi-mode phone or cell phone, or it sends you an IM because you�re in a meeting.�

�Ultimately, it�s all about presence,� says McHugh. �What IP address are you at? What devices are there around you? How can they be exposed in an application? How are they made transparent to an application? The guy who built the service doesn�t want to deal with all those things. It�s exposed APIs or Web services that enable session establishment. That kind of easy service development is our play, our vision for moving into IP and converged services.�

Covad (News - Alert) Communications

Perhaps the best-known of the CLECs, Covad (www.covad.com) has been �riding the wave� of improved IP Communications quality and a diversification of services that SMBs increasingly find tempting.

Eric Weiss, Covad�s Chief Marketing Officer, says, �In 2006 Internet telephony and IP Communications really graduated to Business Class. Some examples of how we tried to contribute to that would be that we rolled out and scaled things such as voice prioritization and the ability to prioritize real-time applications over all of the other data-type applications. With the YouTubes of the world out there taking a lot of bandwidth, we�ve got to make sure that the real-time business class applications such as VoIP still maintain their business �grade�. So, we have a technology and service called VOA [Voice-Optimized Access] that gives you voice priority over data, and we have a number of features available through our Covad Dashboard and our Portal that enable a whole suite of applications such as Find Me/Follow Me, Unified Communications, Unified Messaging, Rich Conferencing, and so forth. We enable those features bundled with the voice.�

�This, in turn, speaks to where things are going in 2007,� says Weiss, �which is not to market these solutions as �VoIP�, but to market them as IP Communications. It�s integrated communications, and the fact that it�s called VoIP just means that IP enables all of the features; take for example, what we put in our Dashboard and Portal.�

Weiss elaborates: �So, in 2006 we graduated to Business Class with those features, with QoS as per VOA, but I still think that 2006 was primarily about the well-funded enterprise users, the early-adopters and the people either moving or doing greenfield deployments. That�s what really characterized IP Communications in 2006. That�s going to transition in 2007 where we�ll basically be done with the early adopters. We�ll stop calling it, marketing and selling it as VoIP, and instead we�ll start marketing it as what is really is: as communications, as collaboration. That may sound simple and �vanilla�, but that�s what the broader market of SMBs wants and needs. And we�re going after that market.�

�Business Class IP Communications is now a better alternative � even for the installed base � in terms of cost, productivity gains, the features, and the need to replace Y2K PBXs, and not just for moves, greenfields and early adopters,� says Weiss. �Our focus on SMBs drives us toward providing tools that help people sell more widgets, bill more hours, see more patients, and so forth.�

�We found that SMBs don�t tend to buy technology,� says Weiss, �but they do want communications to help them. If they�re a law firm and their top customer calls them, they want to be able to have that client find them wherever they are and similarly automatically block a caller who�s merely trying to sell them more paper clips. They want to grow their business, and the only way they can do it is with advanced tools, which is what we�ve built into our offerings.�

�So, that was the primary transition: from early adopters to practical users who are not buying technology � namely, VoIP � they�re actually buying �communications� that helps them get their job done and grow their business,� says Weiss.

�Also, in 2006 there was a lot of confusion in the market about �who to go to for what�,� says Weiss. �There was no category leader in the SMB communications market. Vonage was out there, as well as Skype, and cable companies, traditional telcos, PBX (News - Alert) manufacturers who even sold in the retail chain, and then competitive providers such as Covad. The buyers were pretty confused about who to approach. That will change in 2007. There will be a clear category leader, particularly in SMB VoIP and IP Communications. To be such a category leader, the company must exist nation-wide, so that as SMBs grow and establish branch offices they can be served by the nationwide option. Second, the category leader must offer services or technology with high QoS. Third, it must be an innovative company, which generally doesn�t describe phone companies. Fourth, it must be flexible enough to address the needs of SMBs. In fact, it should preferably be someone focused on SMBs, since it�s difficult to deal with both SMBs and enterprises, and that means that you�re not actually focused. Fifth, it should support both a hosted model and a model that allows for integration with customer premises PBX equipment. Both models are valid.�

�There will be category leaders in each market segment,� says Weiss. �Obviously, we at Covad think we�ll be the category leader in SMB VoIP. We�re already the Number One provider in hosted PBX solutions and we think we�re well-positioned. That�s good for us because there will be a shakeout � more mergers and acquisitions. It�s hard to make a transition from wholesale to direct sales. One very large wholesaler that just made a very large acquisition talked recently about how they�re now moving into the SMB market. That�s easier said that done. We now have close to 40 percent of our business derived from the direct Covad-branded service for SMBs, and to achieve that took a great deal of time and effort � we�ve been working on it for two years now. You have to really make an effort to transition your business from wholesale to direct. It�s nontrivial. But we�re over the hump now.�

�Yet another trend appeared that�s also interesting,� says Weiss. �It concerns the channel. In 2006, agents and dealers, for example, really proved that they could handle both the telecom and IT sides of the business. The channel converged just as the technology had converged a couple of years previously. We have several hundred channel partners out there and we�re seeing really strong growth. Our channel partners can deliver on the promise of IP voice and data and combined services. That really positions us well for growth in 2007.�

Covergence

Rod Hodgman, VP of Marketing at Covergence (www.covergence.com ) says, �I expect to see a shift in the VoIP marketplace in 2007 as users begin to move beyond VoIP to other real-time services. Today�s VoIP market could be characterized as single service � voice � with the user experience being delivered over a 50-year-old TDM phone. But 2007 will mark the beginning of a significant transformation. Service providers are testing pre-IMS deployments that deliver multimedia services � voice, video, IM, and presence � to any device that supports the SIP protocol. This means they will face new challenges in the areas of scaling access, securing the service and managing the environment.�

�At the subscriber edge,� says Hodgman, �service providers will need to be able to predictably scale hundreds of thousands to millions of active endpoints. They will also need to scale phone registrations and fully encrypted, validated and authenticated user connections without impacting performance.�

�Security will also get significant attention in 2007,� says Hodgman. �The SIP protocol is the standard for access to real-time services and as it proliferates in clients, softphones, handsets and mobile devices, like all other IP protocols, it will come under increasing attack. Service providers will have to ensure the authenticity, confidentiality, and integrity of subscriber communications, and they will have to defend against attacks and compromises such as DOS attacks, service theft and others.�

Covergence offers a new kind of dedicated appliance for VoIP service providers and their SIP networks. Called Eclipse, it connects to SIP-based VoIP and real-time services (e.g., voice, video, instant messaging), providing a unified security and management solution for SIP at the subscriber�s edge of the carrier�s network. Eclipse sits between the subscribers and the service provider and guards the network from attacks similar to those that are often seen with other application-layer Internet protocols, such as HTTP and SMTP, but are only now starting to appear with some frequency with SIP. Eclipse can also handle signaling and real-time media streams separately.

Aside from security, Eclipse can also manage inter-service the interactions and upports application interoperability. Eclipse transcodes among various vendors� SIP/SIMPLE dialects and allows for cross-domain presence visibility and messaging connectivity. For example, with Eclipse, providers can achieve interoperability between IBM Lotus Sametime and Microsoft (News - Alert)�s Live Communications Server (the latest incarnation of which is now called the Office Communications Server). Eclipse enables message and presence sharing between such different applications. Moreover, Eclipse enables administrators to define security, control and monitoring policies enforceable on all cross-domain collaboration traffic.

�SIP is the prime component of IMS,� says Hodgman, �and so we�ll be seeing some actual deployments of IMS in carriers and service providers next year and beyond.�

Dialexia Communications (News - Alert)

Ahmed Aina is CEO of Montreal-based Dialexia Communications (www.dialexia.com), which provides advanced voice and data over IP switches for small carriers, ITSPs, phone retail outlets and for VoIP telephone systems for SMBs,� Aina says. �Over the past year, we have seen that combined with the globalization of the economy, the markets, and the deregulation of international communication, the workmanship of digital network technologies in IP shows us that the epoch of telecommunications monopolies has past. The future now belongs to more open and approachable digital technologies.�

�The convergence of voice, video, and data over Internet Protocol is now creating a market for tomorrow�s telecoms valued at several hundred billions of dollars,� says Aina. �Barriers to participation have fallen and the big players are now positioning themselves over the new IP-based digital telecommunications network. There are higher risks if they don�t offer added value and clients are more and more demanding and require quality service.�

�It is now a question of when IP telephony will be adopted, since the key issue today is when to implement and how to manage the migration to converged networks,� says Aina. �The reasons for rise in demand are: the emergence of business VoIP as a new and viable solution; the massive growth of the Chinese telecoms market and developing countries; the purchasing cycle to replace �Year 2000� or �Y2K� time-expired equipment; and lower costs of VoIP calls. All these factors have been expanding market growth into the next year.�

�In the past year there have been world-class achievements in the convergence of voice, video, IM and data,� says Aina. �VoIP development has shown steady growth and progress, especially in the demand for Hosted IP PBXs versus the premises-based IP PBX market.�

�In 2007 larger enterprises will show increasing interest in Hosted options,� says Aina, �since they can support more users without an investment in equipment and IT staff. There will also be a marked growth in the momentum of on-site IP PBXs, which appeal to small and medium-sized businesses due to their advanced features, such as call logs, voicemail, find me, follow-me, and conferencing. The outcome will be that companies will continue transitioning from their legacy TDM systems to fully embrace VoIP technology. What we envision in the near future for IP Communications is a new direction in enterprises moving from do-it-yourself software installations on their own hardware, to purchasing pre-loaded IP PBX appliances. This will become an emerging trend, since configuration, deployment, maintenance and support are much easier to achieve successfully using these high-performance, reliable and easy-to-use turnkey devices.�

�In the past five years there have been many players, but in 2007 we will see a big shakeup of the industry, and only the most robust and solid products and companies will remain,� says Aina.

Dialogic Corporation

After ten months of searching and (finally) negotiations between Intel (News - Alert) and Eicon � which resulted in a sale of assets to Eicon and a name change by Eicon � the legendary computer telephony board maker Dialogic (www.dialogic.com) made its reappearance in 2006. I sincerely hope they will once again throw the same kind of spectacular parties for editors that they did when Howard Bubb ran the company!

In the meantime, Yours Truly asked Dialogic (News - Alert)�s VP of Product Management and Planning, Jim Machi, how IP Communications has progressed over 2006.

�Very well,� Machi responds. �In fact, I�m sure there are many consumers out there using IP Telephony and they don�t even know it. It�s entered the mainstream. Continued improvements in availability, reliability, and redundancy in both the enterprise and service provider space coupled with enhanced voice coders have dramatically improved users� quality of service [QoS] and overall experience. Solutions addressing E911 and Security are abundant. Increased convergence driven by broadband everywhere has increased potential consumers. And lastly, the Session Initiation Protocol (News - Alert) [SIP] is and will continue to play an enormous role in IP telephony networks. Standardization is the obvious benefit but, more importantly, SIP has moved forward real-time multimedia: voice, video, and most importantly Web collaboration solutions.�

�2007 will bring innovation in the application space as providers need to �subsidize� falling voice revenues,� says Machi, �so look for more enhanced services driven by IP such as presence and collaboration solutions. Hosting will gain even more ground as more robust and �easier to deploy� telephony hosting solutions provide an attractive alternative to CPE equipment. Providers will also move toward IMS or IMS-like platforms realizing reductions in total cost of ownership.�

Interactive Intelligence (News - Alert)

�Large scale product adoption is always sparked by some disruptive technology or product introduction,� says Joe Staples, Senior VP of Marketing for Interactive Intelligence (www.inin.com) a company that�s familiar with disruptive product introductions, as it knocked the computer telephony industry (and especially the old PBX, ACD, and VRU vendors) for a loop way back in 1996 when it unveiled its Enterprise Interaction Center (EIC), a comprehensive LAN-based communications system that acts not only as the PBX, but also provides automatic call distribution (ACD), interactive voice response (IVR), fax services, a Java-based workflow engine, and voicemail integrated with Microsoft Exchange.

�VoIP is that disruptive technology that, over the past year, has gained significant momentum and caused companies to reevaluate the TDM-based systems that they deployed years ago,� says Staples. �It has caused them to look at what else they can do, how they can compete more effectively, and how they can improve efficiencies. That reevaluation of their telephony systems, again initiated by the introduction of VoIP, has turned into a strong increase in purchases. We�re in a dramatic upward cycle right now. VoIP, SIP, and the associated applications have given customers a reason to buy. I also think we are at the beginning of the cycle and that we will continue to see strong growth for the next several years.�

Interactive Intelligence should know. Their year-over-year revenues were up 46% last quarter.

Inter-Tel (News - Alert)

Jeff Ford, CTO and President, Integrated Systems, of Inter-Tel (www.inter-tel.com), has been with company since 1983, his senior year at Arizona State University. He has seen many trends in telecom come and go.

�Clearly, the last half of 2005 and coming into 2006, we saw what Rich Tehrani called �first-generation VoIP� capabilities,� says Ford, �which is basically VoIP as a new technology � compared with TDM � for transporting voice. It clearly matured out there in the marketplace. We�ve gone from dealing with the early adopters to general and mature acceptance of the technology and that first-generation VoIP capability. We continue to see accelerating adoption of VoIP.�

�Inter-Tel traditionally focuses on the less than 500 user per site market,� says Ford. �We�re clearly selling to the SMB space. If you take a look at the deployment of IP in the SMB space, it�s generally 10 or 15 points behind the overall industry, but it�s tracking and increasing every day, and the technology has clearly met mainstream capabilities. When people buy new systems, they�re now expecting them to be IP-capable. The majority of SMBs are still purchasing digital phone capability, but they want to have an IP-capable system and they�re deploying some IP features and capabilities.�

�Over the last couple of years, Inter-Tel has pushed forward with its collaboration and presence capabilities,� says Ford, �and that has reflected nicely what�s been happening in the market. We introduced our Unified Communicator product in 2003, which is the hub of our presence management technology in collaboration and then we�ve introduced Web conferencing and desktop video capabilities, integrated with that same platform over the last year. We see these capabilities as the foundation for the beginnings of second-generation VoIP systems that are just now being introduced. These rely heavily on presence, collaboration, SIP and open standards. Instead of just talking about voice transport such as VoIP, we can now talk about business-enhancing applications enabled by various IP applications including VoIP. These will really help drive business value, helping to increase revenues or decrease costs, improve efficiencies and improve customer service.�

�We see 2007 as the year that second-generation technologies and systems really take hold and move into the mainstream on their own,� says Ford, �where customers are looking for open standards platforms with integrated presence management, instant messaging, and other business-enhancing applications.�

Ford elaborates: �After all, everybody for the past several years has talked about the fact the real value of VoIP is not in how you transmit the voice, but instead are the applications that get IP-enabled. We�re biased, I�m sure, but we feel that we�ve been delivering many of those applications over the last several years and we�ve been leading in the introduction of presence-based technologies and integrated messaging capabilities, presence management, advanced call routing based on real-time presence capabilities, desktop sharing, application sharing, collaboration, desktop video, and delivering a richer multimedia environment with our Unified Communicator and collaboration technology, along with our Axxess solutions.�

�Our Inter-Tel Axxess 7000 product will be released very shortly,� says Ford, �and that has many of those capabilities built right into the product�s core. The 7000 is at its core a SIP softswitch so it�s an industry-standard SIP offering that supports SIP-B [which has the benefits of SIP with the feature functionality of MGCP] and B-TXML [Basic Telephony Extended Markup Language] industry standards for phone control. The 7000 also has presence capabilities, collaboration, secure IM, and desktop video, all built right into the offering�s core, as opposed to being sold as an optional �box�. Once again, we feel the 7000 product is out in front in terms of introducing these features. In 2007 we expect to see market acceptance of these features to such an extent that the public will expect them to be an integral part of communications products.�

Juniper Networks (News - Alert)

Juniper Networks (www.junipernetworks.com) has played heavily in the service provider space and its equipment transports all sorts of traffic, including IP telephony, over the world�s networks. Back in 2004, Juniper Networks started to enter the enterprise space, and one of the key things they did was to acquire NetScreen, and subsequently they made other acquisitions, such as Peribit Networks, a developer of WAN optimization technology, and application front-end vendor Redline Networks that operated in the applications acceleration space. They also acquired Funk Software for identity management and access control capabilities.

Juniper�s Senior Director of Product Marketing, Stephen Philip, says, �Over the last year or so, Juniper has maintained its focus on service providers but it has also embarked on an enterprise strategy. Most elements of that strategy are interesting from an overall point of view, but some elements are interesting from an IP telephony point of view. Recently, for example, Juniper has announced a strategy to go after the branch office market. We�re also announcing some development plans and some initiatives for this space. For example, we�re rolling out new versions of the J-series, Juniper�s line of access routers, in particular the J4350 and J6350 enterprise routers. Of course, �branch office� means different things to different people. It ranges among the micro-branch, telecommuters, and small office, up to the branch and regional office.�

�Throughout the summer of 2006 we also rolled out a new portfolio of security appliances,� says Philip �and as we roll out the new platforms, we�re leveraging some of the technology between them. The branch solutions include 11 new Secure Services Gateways [SSGs] security platforms with integrated branch routing and full Unified Threat Management [UTM] capabilities and the new J4350 and J6350 enterprise routers that leverage technology from the Secure Services Gateway (News - Alert) 500-series platforms. Our Secure Services Gateway Platform is an integrated security appliance with a firewall and VPN connecting back to central headquarters.�

�There�s also some additional management work that we�re doing, in terms of developing an integrated suite of management applications to control the systems in the enterprise space,� says Philip. �And also some customer service initiatives to aid enterprise transitions.�

Sanjay Beri, Juniper�s Director of Product Management, says: �One technology you�ll be hearing a lot more of in 2007 are �Application Accelerator� WAN optimization platforms, such as our Juniper WXC 250. Application accelerators are a scalable approach to accelerating application delivery over the WAN. It helps businesses improve application response times, maximize WAN investments, and control and prioritize key applications. Ultimately, the WXC WAN optimization capabilities will be integrated into the J4350 and J6350 routers to provide compression and caching, TCP and application-specific acceleration, and visibility and reporting functions. The integration will be achieved via a WXC services module to leverage the core functions of the J4350 and J6350 to provide an integrated set of QoS, bandwidth management and multi-path functions. Standalone branch J-Series and WXC platforms will continue to be available into 2007.�

NexTone

Nick Turner is VP of Product Marketing for NexTone (www.nextone.com), makers of the IntelliConnect� System that includes such edge devices as the NexTone SBC (Session Border Controller), NexTone MSX (Multiprotocol Session Exchange platform for interconnecting SIP and H.323 networks), and the NexTone IMX (IP Multimedia Exchange platform for interconnecting IP and IMS networks).

Turner says, �We�ve seen tremendous growth, in terms of the number of carriers deploying VoIP, the �amount� of VoIP that they�re handling, and the increasing awareness of the importance of SBC-type products in their networks and service deployments. If you look at the market numbers, they tend to confirm that the SBC space is the fastest-growing in the sector. And in the future, we expect not only continued growth but perhaps an increase in growth as SBCs or SBGs � session border gateways, to use the IMS parlance � becomes more of a known quantity and necessary element to network design.�

�The question arises whether SBCs and SBGs will retain their essential character or morph into something else,� says Turner. �If you look at the MSF GMI [MultiService Forum (News - Alert) Global MSF Interoperability Event] press conference, the participants all highlighted the fundamental need for a session border gateway to provide security and interworking at the boundary points of an IMS network. So its need is becoming �better comprehended� instead of minimized.�

�The terminology itself will also probably evolve,� says Turner. �If you look at the 3GPP or ETSI TISPAN framework, you see lots of delightful acronyms: BGF [Border Gateway Function], BCF [Border Control Function] and new ones evolving such as I-SBC [Interconnect Session Border Controller]. So as we get closer to field deployments and larger scale deployments, the terminology will evolve as real world issues challenge the original design assumptions of IMS. A good example might be the I-SBC; in an IMS architecture document, the I-SBC is made up of functional components such as IBGF, IBCF and IWF. The clearly laid out functions are, from ETSI TISPAN�s perspective, things that can be consolidated into a single product implementation. Thus, it is an �implementation� of multiple ETSI TISPAN functional capabilities. Those acronyms can be considered as just functions, and now we�re seeing new product implementations arise that are organizations of those specific functions and features. A single product can span a number of functions.�

�I think that 3GPP and ETSI TISPAN are clear as to what a function is versus a product,� says Turner, �and again, as we get closer to field deployments and larger scale deployments, we see real-world feedback help drive order and product implementations.�

�On the OSS [Operations Support System] side,� says Turner, �the reason why NexTone developed RSM [Real-time Session Manager, which acts as a centralized policy manager and enforcement point to manage network resources and optimize call distribution] was to deal with the reality of real-time feedback on network performance at a traffic or call flow level, not at the device level, because devices aren�t really relevant. This occurs in real time so that when something breaks � and in telecom, things break � the network operators are made aware and can perform either measured manual intervention or automated intervention to adapt around network problems.�

�What the SBC vendors have proved that there are many difficult performance issues for any type of network element at the network boundary,� says Turner. �The SBC vendors in this space have proven an ability to process large transaction loads, both signaling and media. They�ve got a good two-year lead on other types of vendors in those respects.�

�There are also new architectures and implementations being discussed actively,� explains Turner. �For example, there�s the basic concept of �dissociated architectures� where an existing signaling entity, such as a softswitch, can control a media-only SBC. That�s a credible architecture and NexTone has adopted that from the beginning. We adopted and implemented a dissociated architecture when we shipped our first product. But this leads us back to the same performance problems we have in all types of networks, certainly voice networks, where calls per second and busy hour call attempts are an issue. SBCs or whatever controls them must deal with that at scale. So we look at various labs, such as CT-Labs, where they can measure the performance of an SBC, and you still see that terrible legacy come back at us, which is: how many calls per second can be handled? Under duress such as a denial-of-service attack or just in plain sailing on the network, SBCs in general, and NexTone in particular, has proven effective at handling high transaction loads. We�ll continue to keep pace with future network growth.�

�The other unique attribute of the SBC space in general and NexTone in particular, is that we�ve been sweating out problems in the field that weren�t anticipated in abstract, whiteboard environments, and which are necessary to solve in order to deliver a production network for a carrier,� says Turner.

�Security is increasingly allied with SBCs instead of separate firewall boxes,� says Turner. �SBCs have become the nexus for certain kinds of security approaches. If we go back a few years, SBCs were considered a nominal space with limited time to live. But what happened in 2006 is that SBCs have become both a known commodity and an understood and appreciated function in the network. It�s a core competency area.�

�The shape, form, and future of SBCs are open to great debate and discussion,� says Turner. �What will prevail is the real-world experience aspect � we�re dealing with a new type of network architecture. When we look at IMS, it�s a wonderful design and it certainly has its specific benefits, but those abstract concepts have yet to meet sufficient field-level experience for anyone to know what will break in specific scenarios. SBCs have been, to put it mildly, tried and tested. And that�s been the reason for their significant core competency development. They will continue to flourish in 2007.�

Nortel (News - Alert) Networks

Phil Edholm, Chief Technologist and VP of Network Architecture at mighty Nortel (www.nortel.com) says, �From my perspective, 2006 was a significant year in about three different ways. From a company perspective, we reached the turning point where the volume of VoIP and related forms of IP Communications are now the majority of our business. That was a big transition. Obviously, a major occurrence in the industry has been the adoption of VoIP and the challenges in some customer environments of actually finding the value proposition in the world beyond the proven TDM environment. This year, we saw that transition happening.�

�The second major IP Communications �impact of the year� was the ICA [Innovative Communications Alliance] announcement,� says Edholm. �In July 2006, Microsoft and Nortel announced a strategic alliance at technology, marketing, and business levels involving a shared vision of unified communications. [Editor�s Note: The two companies formed the Innovative Communications Alliance (www.innovativecommunicationsalliance.com) as a �go-to-market� vehicle.] There�s been a transition happening of communications moving from being a separable environment to being part of a collaborative space. In a collaborative space, you have to bring three elements into play: First, management of documents, content, information � the things that you�re going to collaborate about; second, the integration of workflow � when you�re going to collaborate and what�s going to bring you together to collaborate; and third, the modalities of communication which may range from instant messaging [IM] to voice and video in real time, but including other things too.�

�By bringing these three things together, one can create an environment for this transformation around collaboration,� says Edholm. �There�s also the recognition that this is a new and different marketplace. It�s actually changing the definition of telephony from a voice communication system to making it a part of a larger collaboration environment for knowledge workers, who are obviously a small percentage of today�s workforce, but a growing percentage.�

�So, from our perspective, the ICA announcement with Microsoft is a recognition by Nortel that this collaborative knowledge worker unified communication environment is going to become a significant force in the industry,� says Elholm, �and it would have two impacts: it will have a significant impact on many companies in terms of how they build and integrate their telephony environments over the next several years, but it will also affect companies that are predominantly staffed by knowledge workers, such as consulting companies. This will become the standard paradigm of communication in those environments.�

�Nortel recognizes this,� says Edholm, �and Microsoft recognizes that if you want to build a collaborative suite into your products that handles document management, workflow and the extended workflow environment through their relationship with SAP (News - Alert), then it�s very difficult to do it alone without working with a company, such as Nortel, that has expertise in real-time communications. So I think the ICA announcement was in many ways the pivotal event of 2006 in terms of redefining the market space from being an IP Communications marketplace to being an IP Communications environment integrated with these other collaborative tools in a whole new transformation of the market.�

�The third major area is the path over which we�re traveling,� says Edholm, �with our announced direction and changes in our product portfolio concerning our use of conferencing as the underlying communications paradigm of our applications suite. So, instead of having conferencing being an adjunct to communications, it becomes the basis of communications. Instead of me making a connection between my device and your device, we actually base the communication on joining via a conference facility. You could implement that in a call center, for example, where you bring the customer in and tether them in a conference with a customer service representative. It�s the first presaging of this next transformation in communications, where conferencing becomes the fundamental communications paradigm.�

�What we�ve seen this year to back up that idea is that we at Nortel sold a large number of VoIP conferencing systems as part of our VoIP offer,� says Edholm. �Those sales resulted in an explosion in two-party conferences, because people realize that the probability of a meeting occurring if I say, �I�ll meet you in a conference� is much higher than if I say, �I�ll call you�, but when I do call you you�re on the phone and then you call me back and I�m on the phone and so the meeting never happens. The conference paradigm is, thus, a way of not guaranteeing but certainly increasing the probability of a meeting actually occurring.�

Edholm sums it up: �So, to recap, these are the three predictors of where we�re going: The fact that VoIP has moved from being a kind of data curiosity to the predominant way that real business telephony is deployed. The second is the whole conferencing arena, but the single biggest event in the industry this year was the debut of the ICA, which heralds the transformation of IP Communications from a standalone technology to being integrated into a true application structure through document management, workflow management and integration, by working with companies such as Microsoft. That�s really going to be a major transformation.�

pbxnsip (News - Alert), Inc.

�The past year for selling software based IP PBXs has seen a lot more successful deployments according to Kevin Moroz, VP of Sales at pbxnsip (www.pbxnsip.com), makers of a virtual PBX that can connect directly to an ITSP or a customer premises PSTN gateway.

�The market is yearning and searching for robust, feature rich, easy-to-use systems,� says Moroz. �We have had thousands of downloads of our product this year and obviously expect even more next year. The lab trials and viability phase is over and now the widespread deployments are taking root. We are finally seeing multiple sites and 100+ user phone deployments become more commonplace. The hosted PBX market is bubbling up even more as small companies all want to be the next Vonage and are deploying systems all over the world as quickly as possible. The key to more explosive growth is keeping SIP highly interoperable. The market loves to be able to pick and choose their IP phones and gateways as well as their own Internet Telephony Service Providers [ITSPs]. We need to keep educating the market place and �eating our own dog food� as we say, to keep the market growing.�

SpectraLink

2006 also saw new inroads by WiFi phones and the beginnings of dual and triple mode technology. SpectraLink (www.spectralink.com), for example, specializes in workplace WiFi wireless telephony, offering seamless integration of wireless VoIP and traditional telephony platforms.

SpectraLink�s Vice President of Marketing, Ben Guderian, says, �Looking at this past year, there were things certainly occurring specific to the wireless space. For example, the 802.11e standard for WiFi quality of service [QoS] was finally ratified. People were happy that it was done, though it didn�t have a huge impact in terms of what we�re doing since we (and others) had already dealt with finding ways of doing high quality IP telephony over wireless.�

�Much of the energy that we put into this that�s unique to us in the wireless space has to do with the compatibility issues involving the various access points that are out there; this adds a whole other layer of complexity on top of what everybody else has to deal with in the wireless IP telephony world,� says Guderian.

�The other thing that we�ve moved toward and that we�ll see a lot more of in 2007 is more support for many of the switch platforms that are running SIP,� says Guderian. �To date, SpectraLink has primarily focused on large enterprise customers and much of what we do goes into TDM switches, and a lot of the stuff we sell goes through our OEM partners that have their proprietary IP protocols, companies such as Nortel, Avaya, NEC (News - Alert), Alcatel, and so forth. We even support the proprietary offerings from vendors such as Cisco and Mitel.�

�This past year Cisco came out very strong, talking about support for SIP, along with others,� says Guderian. �So we will most likely see a shift � not overnight but over time � moving away from proprietary protocols such as Cisco�s SCCP or Skinny Client Control Protocol [used between Cisco Call Manager and Cisco VoIP phones] into more SIP-based technologies. Of course, everybody�s going to have their own flavors of SIP-enhanced extensions too. That�s certainly where we�re seeing things moving, and we�re taking that into consideration as we develop new products.�

�Also, this year we did some prototyping and some demonstrations of some softphone applications as a means to show how you could get a third-party wireless device to work with not only an IP PBX, but more importantly, with TDM technologies,� says Guderian. �I realize that, in the context of Internet telephony and VoIP, this may appear less interesting, but it certainly got the attention of many customers who are trying to squeeze a few more years of useful life out of their TDM switches. It leverages what we do in terms of making the client device more of a thin client and taking advantage of the server or gateway we provide to do the PBX integration.�

�We�re not trying to force anybody to go in any direction and so what really drives us is making sure that we�re compatible with what the customers are trying to use, and sometimes that makes us a bit �old school�, because we still maintain a great deal of TDM support,� says Guderian. �The reality is that the markets we serve are still populated with TDM switches.�

�There are many cell phones and PDAs out there today that support both cellular and WiFi,� says Guderian. �These are the kinds of devices to which we�re targeting our softphone, but, interestingly enough, many of these devices do a very bad job of supporting voice on the WiFi side. It�s getting better, because everyone�s starting to realize that WiFi radio might be useful for people wanting to use an IP telephony service � perhaps even Skype � from home or a hot spot. Generally, those devices have required a fairly high knowledge of how everything goes together in order to make them work correctly. Again, a lot of the IP telephony things out there today require a bit more knowledge than the average consumer has to get them to work. And then you layer on top of that the wireless piece, all kinds of variability, and issues you must deal with, such as whether it will work okay on your home network, corporate network, or a T-Mobile (News - Alert) hot spot. There are a lot of �interesting� things that can happen because you�re faced with different levels of security, and different user authentication things are going on. So the issues go well beyond making the voice stuff work over WiFi, and then over the broadband connection; sometimes it has to do with just getting the device up and running on the network.�

�In 2007 we�ll see more Voice over WiFi used in places where the company is comfortable with it and it�s mission critical,� says Guderian. �The best examples of that are the big home improvement chains, both Lowe�s and Home Depot have pretty much standardized on the technology. You can add to that many big hospitals and manufacturing plants. Many of the early adopters of WiFi and people who installed wireless LANs to help do other things, such as barcode scanning, are way up front in supporting voice on the networks. However, Voice over WLAN is still very much in its early stages as a consumer play, just because it isn�t really plug and play yet. In the enterprise, you need a pretty savvy IT staff today to make it all work. We make it as plug and play as possible because we provide all of that extra PBX integration and so forth. But what we do doesn�t necessarily translate that well over to the consumer world, at least not yet.�

Sylantro Systems (News - Alert)

Sylantro (www.sylantro.com) is a leading provider of software used by service providers to deliver hosted VoIP applications and services for business, consumer, and wireless customers.

Syltantro�s Senior VP of Global Marketing, Ron Raffensperger, says, �From Sylantro�s standpoint, a couple of interesting things happened this year. For example, IMS started to become real. It�s beginning to move from the labs into early trials. Also, consumer VoIP really started to hit the mainstream this year. We�ve seen huge growth in acceptance from several of our customers.�

�Also, we started to see examples of seamless mobility this year,� says Raffensperger, �but I don�t think it will really get going until next year. Some of that is sorting out, involving the availability of dual mode handsets, the availability of full PBX functionality on a mobile phone. You got a hint of them this year. I think they�ll be major players next year.�

�Will session border controllers retain their identity in an IMS world?� asks Raffensperger, �I think they�ll pretty much stay the same. There are a few little twists, but generally speaking SBCs will continue to be a fact of life just because the carriers need to protect their networks.�

�There are a number of speed bumps on the road to IMS,� says Raffensperger. �We�ll see them in mobile first, because that�s where IMS was originally defined. And we�re seeing a lot of them in greenfield installations. We�ve got an installation in Pakistan with Motorola (News - Alert) that�s being turned up as we speak, and that�s a great one for IMS because it�s a greenfield � there was no �last mile� because there was no existing infrastructure.�

�One uncertainty we�ll confront in 2007 is whether or not IMS becomes a reference architecture,� says Raffensperger. �Also, the original concept of IMS was that things were going to be plug and play, and service providers could go get a switching CSCF [Call Session Control Function] from one company, a SCIM [Service Capability Interaction Module] from another, a feature server from yet another, and just plug them all together and everything would work perfectly. What we�re actually seeing is that the traditional network equipment providers are going to be the folks that choose and guarantee what works together, and you�ll see fewer service providers trying to pick �one from column A and one from column B� and so forth. Those who own the infrastructure will decide what runs on it. You see that starkly in the Verizon (News - Alert) A-IMS [Advances to IMS] proposals. It also blows up the whole Net Neutrality movement. A-IMS has as its subhead, �not on my network you�re not�!�

�We had a good year,� says Raffensperger. �The keys things have been getting some new partners, and some analysts have mentioned how Nokia has a new business communications service offering that they�re bringing out that includes Sylantro. It�s been an exciting year.�

U4EA (News - Alert) Technologies

Peter Thompson, Chief Scientist at U4EA Technologies (www.u4eatech.com), says, �We have a particular interest in quality of service [QoS], by which we mean actually architecting and engineering a network so that it delivers reliable services. One of the things I�ve noticed in the last year is more interest in QoS in a broader sense. People realize it�s something they should be concerned about. This is a change from the situation a few years ago, when it was quite difficult to get people interested in it.�

�Another interesting angle on this has been the debate over Net Neutrality,� says Thompson. �At one point it looked as if the U.S. Congress might legislate against QoS, which would have been interesting. Yes, one of the bills on the table was that it would be illegal to prioritize one type of traffic over another. The debate became furious, and generated more heat than light. Fortunately, Congress didn�t pass that bill. People tend to forget that prioritizing one type of traffic over another only makes any difference if the area of the network you�re dealing with is congested. The big carrier networks are all full of fiber and DWDM [Dense Wave Division Multiplexing] technology and don�t really get congested. So, even if they had the technology in place to prioritize one type of traffic over another, it wouldn�t make any difference. Given the bandwidths that exist, it�s just irrelevant. The only place it makes any difference is where links are actually congested, which could be on the link to the customer � the access link. But it�s not like the customers are going to be using the services of Google and somebody else at the same time on the same link and then observe that Google is better because they paid more money to get a better prioritization. That�s absurd. The whole thing is a bit of a tempest in a teacup, from my perspective.�

The other Net Neutrality fear is that carriers would block competing VoIP services, which they in fact do on occasion,� says Thompson. �But the act of completely blocking something is part of a completely different argument that has nothing to do with whether or not the carrier bestows different priority treatments to the delivery of competing services.�

�Incidentally, I�m not implying that simply overprovisioning bandwidth is a panacea,� says Thompson. �Historically, people have said that we have more than enough fiber capacity and we�ll never run out of bandwidth. Of course, people said that about 64 Kbps frame relay connections, and they said it again when ATM and SONET reached 155 megabits per second. But, to paraphrase the old adage, �applications expand to steal available bandwidth�. I must admit I have more faith in the ingenuity of people to devise new applications that will exploit the available bandwidth than I have in the ability of companies to engineer entire networks to offer more bandwidth than people can use. So I do believe there will always be a need for QoS. There are a lot of people who would like that need to go away, but it never will.�

�What�s confusing is that many things are called QoS,� says Thompson. �Things like Cisco�s MPLS tend to be less about traffic prioritization and more about traffic route control so the packets avoid areas of congestion in the network to begin with. When you�re working on a network core and you do have plenty of capacity, and you also have lots of alternate routes available, then it makes sense to use MPLS to control the paths that packet streams can take across the network, so that they don�t overload any individual links and cause congestion. In core networks, that�s fine. But when you move out toward the customer, that approach tends to fail: the capacity of the links decrease, and the choice of alternate routes becomes restricted. That�s exactly where some kind of packet prioritization and congestion management makes sense.�

�Different kinds of QoS thus become applicable at different communication layers,� says Thompson. �Our particular specialty is optimizing the queuing and scheduling of packets at Layer 3 in the OSI Communications Model. But at Layer 4 there are things you can do with error correction, packet retransmission, jitter buffering, and that kind of thing, which goes a long way toward smoothing out any problems left over from the communications layer below.�

�Part of the trouble with MPLS is that you don�t tend to get any actual guaranty of what the ultimate performance of the network will be,� says Thompson. �You may get some guaranty of bandwidth, but I don�t think many service providers are at the moment stepping up to provide any kind of guarantees about what level of packet loss, delay or jitter you�ll experience. They�ll prioritize some of your traffic to be �better� but they won�t necessarily give you guarantees as to what they can deliver. In 2007, however, I think we�ll see some of that starting to happen. There will be more gradations of service � MPLS only has three levels of service � and we�ll see more actual guarantees of end-to-end service. Once that starts to happen, it then opens the door for the development of services to real demand, both in terms of stuff like decent video conferencing or services bundles for integrated sectors such as healthcare or whatever.�

�Providers will be able to offer not just the obvious real-time stuff, but also this whole business of the �next wave� of stuff after datacenter consolidation � web services and instances of remote software execution,� says Thompson. �That�s fine, just as long as you can ensure that the network cooperates. Today, some people may experience some difficulty in getting reliable performance, just as many enterprises find it difficult to achieve reliable performance when they centralize apps across a corporate network. By doing this the enterprise gets more and more dependent on the network being able to deliver a predicable level of performance. As long as it can do that, you can engineer the whole thing to respond well enough for the users, but the more that different bits of applications have to be coordinated across the network in order to deliver the result, the more vulnerable you become to having some packet being lost or some delay being large, resulting in unacceptable performance to the end user. But that will change as we see providers deliver SLAs [Service Level Agreements] with some real �bite�.� IT

Richard Grigonis is Executive Editor of TMC�s IP Communications Group.

If you are interested in purchasing reprints of this article (in either print or PDF format), please visit Reprint Management Services online at www.reprintbuyer.com or contact a representative via e-mail at [email protected] or by phone at 800-290-5460.


[RETURN TO THE IT TABLE OF CONTENTS]



Today @ TMC
Upcoming Events
ITEXPO West 2012
October 2- 5, 2012
The Austin Convention Center
Austin, Texas
MSPWorld
The World's Premier Managed Services and Cloud Computing Event
Click for Dates and Locations
Mobility Tech Conference & Expo
October 3- 5, 2012
The Austin Convention Center
Austin, Texas
Cloud Communications Summit
October 3- 5, 2012
The Austin Convention Center
Austin, Texas