Far exceeding the organizer’s expectations, the Lync Conference sold out weeks before the event with over 1,000 paid attendees.  (Based on the badge swapping in the hotel foyer, there must have been another 300 people “around the fringes”.)   The event was an interesting mix of roughly 1/3 end-customers, another 1/3 resellers/partners and 1/3 vendors, analysts and press - all mixed into hours of great sessions, keynotes and exposition space.  What does this mean?  Lync has arrived and it is HOT.  

We at AudioCodes used the event to kick-off our One Voice for Microsoft Lync branding campaign, featuring a new informational video that was hard to miss on our giant video screen.  Feedback from the analysts and customers was very positive – the value of one source for the network elements, services and support is clear.

The AudioCodes 420HD, 430HD and 440HD IP phones were quite the hit with the attendees.  Most notably was a “phone premonition” by the popular blogger Matt Landis foreseeing the arrival our 440HD phone.  (I’m personally pitching to our executives that we informally call our 440HD “the Landis Phone” going forward.)

My session titled “Avoiding the Pilot Trap” was a hit at the show.  Featuring an end-customer presentation from Jeff Bryngelson, Network Manager at American Axle & Manufacturing and a partner perspective from Benjamin Tosado, Principle at Conquest Technology Services, the session discussed some of the challenges that IT Managers and Administrators run into when their Lync pilot projects run into troubles and some first-hand tips on how to avoid a similar fate.  Even the critical consultant Sheila McGee-Smith gave us a “best session of the conference” twitter nod for including real customers to share their stories.  You can view a recording of the session on YouTube.

AudioCodes, Plantronics, Intelepeer and ScanSource Communications collaborated to organize the first ever “Voice LyncUP”, a social event that cruised the San Diego harbor on Tuesday evening.  The event brought together close to 300 end-customer, partners and sponsors out on a cool and windy evening for an opportunity to socialize and share their Lync experiences.   We’re looking forward to a bigger and better event next year with more sponsors and attendees.

And most importantly, the event gave us an opportunity to meet, collaborate and listen to a number of our current and prospective customers.  Having real eye-to-eye discussions about the needs and wants for business adoption of Lync is impossible to put into words.   We left with a new appreciation of what has been accomplished and what is to come from some of the early adopters and highly innovative community that makes up the Lync ecosystem.

One final thought – for those of us that attended this first-ever Lync Conference event, I suspect there will be a day in the future when we say “I was there when….”   Looking forward to next year.

You can reach Alan at:  alan.percy@audiocodes.com

• TDM PBXs would use a media gateway with the TDM side facing the PBX - in most cases emulating the network side of a PRI circuit.  
• For IP-PBXs, an E-SBC would be used in a Back-to-Back User Agent (B2BUA) mode that terminates the Skype session, performs security checks and then re-initiates the call into your IP-PBX in a format that it understands.

As a result of these issues and the need to integrate SIP-based communications systems with a wide range of SIP Trunking service providers, a whole new category of customer premise equipment has recently evolved--the Enterprise Session Border Controller (E-SBC). The E-SBC is designed to be located on the customer premise and sit between the Local Area Network and the external Wide Area Network. Unlike the larger and more complex carrier-oriented Session Border Controller (SBC), the E-SBC is "right-sized" for a range of medium and large enterprises.

Unique functions of an E-SBC include:

Security: Often the first attribute to get mentioned about any SBC. Unlike a firewall, both carrier-class and enterprise-SBCs operate at OSI layers 3 and 4, interpreting the SIP messages and using the information gleaned from the session negotiation, to make intelligent decisions about which request is valid and which message is part of an attack. E-SBCs offer a "front guard" that protects the business network from possible attacks that originate from outside the business (the Internet), elsewhere on the WAN (the carrier) or within the business (an inside job). Stateful packet inspection, Access Control Lists, Topology hiding and Application Layer Firewall functions help keep the bad guys out and let the trusted parties in. Other facets of security include encryption--allowing the SIP sessions outside the business to be fully encrypted without the cost of having encryption on every device or system within the network.

Interoperability: sometimes forgotten, but equally important is the ability to integrate different SIP-based systems from different vendors or vintages with a range of SIP Trunking carriers. As a result of the wide range of protocol options within the RFC-3261 SIP specification, two systems can be completely within specification, but unable to communicate. SIP mediation is often required to convert from one vendor's version of SIP to another. This is especially important as larger enterprises integrate numerous different SIP systems together due to acquisitions, or that may have been bought at different times. An E-SBC eliminates this issue by implementing a back-to-back user agent, essentially terminating one SIP session (using one set of rules) and establishing another session (with a different set of rules), interconnecting previously incompatible systems. Having an interoperability solution is key in maintaining choice for the business and eliminating vendor "lock-in" commonly found with large "standard, but closed" communications systems. 

NAT Traversal: one of the big benefits of SIP based communications systems is the ability to put phones in employees' home offices, hotel rooms, etc. for anywhere-anytime communications. To do this, the communications system must be able to traverse the Network Address Translation (NAT) function found at the far end--a built feature found in most home and small business routers. Enabling the remote phones and workstations requires logic to deal with the IP address changes and port number re-assignments that are the result of the far-end router NAT.

Survivability: this is a new twist that some early market trials identified as an issue with decision makers. Basically, buyers need a "CYA" or back-up plan that will allow the business to continue to operate if there are issues with the SIP Trunks. This may be just during the cut-over period, or part of a longer-term disaster recovery plan. The most logical back-up to SIP Trunks are TDM trunks. Not necessarily a one-for-one backup, but a reduced number of TDM trunks that would be able to stand in and allow for emergency or limited service calling. Until recently, this required a separate media gateway that was installed next to the E-SBC and a SIP Proxy to make decisions about when and where to direct the traffic to the TDM trunks. Fortunately, appliances that combine the three capabilities (E-SBC, Gateway and Proxy) together in one device are now appearing on the market, eliminating the costs and shelf space needed for the three separate devices.

As the adoption of SIP-based IP-PBXs and Unified Communications systems grows along with SIP Trunking, an Enterprise Session Border Controller (E-SBC) will become a common tool that network designers use to protect and interconnect their communications.

This is why the time was right for AudioCodes to launch our E-SBC product line, based on the popular Mediant 800, Mediant 1000 and Mediant 3000 hardware platforms.  

For a more detailed look at the E-SBC and it's role in enterprise communications, view the on-demand webinar titled "SIP Trunking - Security, Survivability and Interoperability"

For more information on the products, see:  www.audiocodes.com/e-sbc

• Reduction of trunk lines to branch offices
• Elimination of toll charges for inter-office calling
• Consolidation of trunking facilities
• Centralized applications (voice mail, IVR and ACD systems)
• Easier remote management and elimination of "truck rolls"
• Improved productivity with Unified Communications capabilities

• Survivability - what happens at the branch if the wide-area network goes down or accidentally cut? An extreme example of this is the outage in San Jose, CA on April 8^th, 2009 where a major fiber optic line was cut by vandals.
• Local Numbers - will you still have the same local numbers that long-established customers have on their prescription bottles or refrigerator magnets?
• E911 - if there is an emergency, will the first responders be directed to the right facility?
• Broadband Availability? - while quite common in urban and suburban areas, wireline broadband is either very difficult to get or very expensive in most of rural America. Common wireless technologies including EVDO, WiMax and satellite are not conducive to voice traffic and may insert significant latency or jitter.

A Fully Distributed architecture with separate stand-alone equipment at each site pushes all the intelligence to the branch offices, but doesn't consolidate resources or save enough money to make it viable over the long haul. Many of the smaller IP-PBX or communications appliance vendors are touting this architecture, mostly because customers are used to this from the key system days.

A Fully Centralized architecture moves all the intelligence to the core with a large IP-PBX or Softswitch at the headquarters or hosting site and just network equipment and IP phones at the branches. This architecture dramatically reduces costs of the equipment at the branches, but is highly dependent on the quality and stability of the WAN. Local number portability and E911 questions create other problems.

It seems that a hybrid architecture that combines the the best attributes of the fully distributed and fully centralized is much more realistic and is becoming the reference for most network designs going forward. This architecture puts some intelligence at the branch offices, which could include either a small IP-PBX or SIP Proxy to handle intra-office calls or emergency calling. The large centralized IP-PBX or Softswitch would manage inter-office traffic. Other application components including voice mail, local IVR and ACD at the branch offices off-load these services from the central site and allow them to operate with less expensive WAN services or in cases of a WAN failure.