Lync Server 2013: Enterprise voice speaks up

The Enterprise Voice functionality of Lync Server 2013 has a number of enhancements from both the administrative and user perspective.

Alan Maddison

As part of the Lync integrated communications platform, Enterprise Voice has always been a robust, enterprise-class Voice over IP (VoIP) solution. While Lync Server 2010 represented a considerable update in terms of voice capabilities, the changes that have been introduced in Lync Server 2013 are even more significant. Besides Enterprise Voice, Lync 2013 has additional new features such as M: N trunk routing, session management, call forwarding and simultaneous ringing (or “simulring”) authorization, and calling number translation.

M: N trunk routing

M: N trunk routing and session management are the two most important capabilities introduced in Lync 2013. The definition of a trunk in Lync has been modified to facilitate these changes. In Lync 2010 and earlier, trunks were simply a connection to and from a gateway, regardless of whether the gateway was to a PBX or a public switched telephone network (PSTN). On the trunk, you could manipulate certain elements, such as translating the outbound called number, by applying normalization rules.

In Lync 2013, however, trunks are defined as “tuples”. These represent the combination of a gateway Fully Qualified Domain Name (FQDN)—or IP address—and port number, plus a Mediation server FQDN and port. Now you can have a many-to-many (M: N) relationship between Mediation servers and gateways.

In practical terms, having an M: N trunk routing relationship means you can have multiple gateways associated with the same Mediation server pool, and gateways associated with multiple Mediation server pools. You can also have gateways to the same Mediation server pool with multiple unique associations.

This change is most obvious in two places within Lync 2013. The first is in Topology Builder. There is now a container named “Trunks”. When you define a gateway, you’ll have to associate it with a trunk. The second place you’ll see it is when working with voice routes. In Lync 2013, a route is now associated with a trunk (not a gateway, as it was in Lync 2010).

This many-to-many relationship is a significant improvement over the 1:1 relationship in Office Communication Server 2007 R2 and the one-to-many (1: N) relationship in Lync 2010. It’s important to remember, though, that this new capability has some potentially significant implications for designing Lync voice-routing topology.

For example, you can now have a true mesh topology within your Lync environment. All your Lync Mediation servers are connected to all your gateways. While Lync 2010 supported the concept of virtual gateways, this new approach adds a great deal of flexibility and simplifies the topology and management for organizations with large numbers of sites.

One of the most effective examples of the benefit of these new trunks is you can now directly connect multiple Mediation servers (pools) to a single centralized Session Initiation Protocol (SIP) trunk. This change alone creates a tremendous improvement in resiliency because it removes a single Mediation server or pool as a point of failure.

In Lync 2010, the SIP trunk Session Border Controller (SBC) was associated with a single Mediation server or pool. If that server or pool failed, then the voice route was down. In Lync 2013, this will no longer be true, as all your other Mediation servers could have a direct routing relationship with the SBC.

Session management

Within the world of unified communications, session management is often defined as the ability to have a single, centralized dial plan to integrate and aggregate endpoints and system connectivity (such as PBX or PSTN connectivity). Within the context of Lync 2013 voice enhancements, there are some welcome basic session management capabilities.

These new session management functions extend the management and routing capabilities of Lync to cover the broader telecommunications environment. As more organizations implement Lync to replace legacy PBXs, these session management capabilities will be essential for the period of coexistence as PBXs are retired.

The session management capability is made possible by the concept of trunks as tuples. By changing the definition of a trunk and enhancing the routing capabilities of Lync 2013, you can now implement inter-trunk routing and use Lync to provide a centralized dial plan for your entire telecommunications infrastructure.

Because each trunk is unique and subject to all of the existing Lync routing capabilities, you can control which PSTN usages and voice routes are assigned to a trunk. You can also control any normalization or number manipulation occurring on that trunk.

From a routing perspective, inter-trunk routing provides additional routes, such as incoming PSTN calls to an IP-PBX system via Lync, outgoing IP-PBX calls to a PSTN network via Lync, and outgoing IP-PBX calls to another IP-PBX system via Lync. Moreover, it will support media bypass on all inter-trunk routing.

Call forwarding and simulring authorization

Call forwarding and simulring authorization are popular features in most Lync deployments, but the lack of granularity in call authorization in earlier versions of Lync presented some challenges, particularly for larger organizations. Fortunately, Lync Server 2013 provides a wider range of configuration options for call forwarding and simulring that address these challenges.

Besides determining which users can forward their calls and simultaneously ring to another device, you can also control how calls are routed using voice policies. There are three new options associated with the voice policy based on call usages. These route to internal Lync numbers only, or route based on a custom usage.

Enhanced 911 (E911): You can now specify a distribution group as a contact point during emergencies. For organizations with security personnel, this will be a welcome change. You can also customize disclaimers based on location policy. Finally, you can force clients to request location updates at specific hourly intervals, instead of the static four hours used in Lync 2010.

Calling number translation: As part of Trunk Configuration in Lync 2010, you could modify the format of the dialed number using an outbound translation rule to ensure the format matched the one required by the gateway. However, you couldn’t do the same with the caller’s number. With Lync 2013, you’ll now have the option to also translate the caller’s number.

Hybrid voice deployments: Hybrid deployments let you use Lync Online to leverage the on-premises Enterprise Voice infrastructure as if remote users were on-premises users. This lets Office 365 users place and receive calls from the on-premises gateway. This topology also supports Media Bypass, E911 and Call Park. You can use Media Bypass when the Lync Online user and gateway are in the same network location. E911 for a Lync Online user works the same way as for an on-premises Lync user. Location information is automatically retrieved by the client and can be transmitted during an emergency call.

Figure 1 and Figure 2 provide a comparison of the supported features for each different topology.

Figure 1 All versions of Lync provide most of the basic telephony functions.

Figure 2 Some of the more-advanced functions are present only in recent versions of Lync.

UX enhancements

While many of the changes to Enterprise Voice enhance deployment and operational capabilities, there are some welcome UX changes as well.

Voice mail escape: While there are many advantages to simulring, there was always the possibility that when ringing to an alternate device, an external voice mail system would answer the call. To prevent this from happening, Lync 2013 lets you configure a timer. If the call is answered after the defined range of time, Lync 2013 determines the call to have been answered by voice mail and disconnects the call. As the other registered endpoints continue to ring, this lets the user answer the call or lets the call be routed to the Exchange Unified Messaging server and sent to corporate voice mail.

Manager and delegate enhancements: There’s now additional flexibility in how a delegate can handle a manager’s incoming calls. Lync 2013 will now let a delegate receive a manager’s incoming calls on any of the numbers he has configured in simulring.

Conference dial-out improvements: Earlier versions of Lync let conference participants receive a call-back from the audio conferencing server regardless of whether the participant was enabled for Enterprise Voice. Lync 2013 makes it easier for you to let users who aren’t enabled for Enterprise Voice initiate a dial-out from a conference. Meeting organizers who use this Conferencing Policy setting can let their participants join the audio portion of a conference through the PSTN.

Response Group Manager role: The Response Group Manager role is reprised in Lync 2013. Once you assign a user to this role, they can modify all Response Group properties, including workflows, queues and groups. This lets your users take control of the Response Groups for which they’re responsible and simplifies your life.

Enterprise Voice is a core capability of Lync that continues to improve with each release. While many organizations have deployed Enterprise Voice and will benefit from the improved capabilities in Lync Server 2013, there are many organizations that have not taken this last step in their Lync deployment. As a robust, enterprise-class VoIP solution, Enterprise Voice should be part of every Lync deployment.

Alan Maddison is an 18-year veteran of the IT industry, focusing primarily on Microsoft technologies. For the last seven years, he has worked as a consultant focusing on delivering professional services. He’s currently a senior consultant with Microsoft Consulting Services.

Related Content

• Lync Server 2013 and Exchange Server 2013: Powerful communicator
• Lync Server 2010: Archiving Essentials
• Unified Communications: Complete Communication Anywhere