How to cut through the complexity of Lync 2013

As an IT admin, I like to think I’m pretty flexible. I like learning new things, and I’m always happy to play with new technology. But I’ve had a hard time with Lync, Micxrosoft’s videoconferencing and telephony server.

Most of us in IT understand the networking side to our world, but the telephony side crushes us. Lync requires understanding all the big foundational Microsoft technologies (Exchange, SharePoint, System Center, Active Directory, and SQL), plus networking savvy (DNS, certificates, and firewalls) plus all the telephony concepts (trunks and routes). To bring all this together into a modern packet-based architecture is tough.

[ J. Peter Bruzzese details the top 7 new features in Lync Server 2013. | Stay abreast of key Microsoft technologies in our Technology: Microsoft newsletter. ]

It’s no wonder that, when confronted with Lync, so many in IT say, “No thanks. My PBX is fine.” No doubt your PBX vendors are happy about that; surely, some even foster misperceptions and uncertainty about Lync to keep IT wed to the PBX.

Despite my frustrations in trying to figure out Lync, I decided to persevere with help from Jamie Stark, Microsoft’s senior technical product manager for Lync. The conversation was quite eye-opening, so I’m sharing its gist here.

For starters, I asked why Lync has a Web-based console (I’m not a fan of Web-based consoles). Stark said it derived from a decision a few years back that PowerShell was going to be integrated into everything. In that context, Microsoft could have gone with the MMC front end or had a Web-based console. The latter is a little easier to get to from any workstation, so it’s more flexible for admins. And the telephony folks, who come from a non-Microsoft background, typically worked on Unix, Cisco, or other appliances that usually have Web-based consoles. One message Microsoft had to reinforce with the telephony folks is that it’s software and it runs on a server — a major shift from the appliance-style devices common in telephony where the hardware and software are intertwined.

I asked Stark about the complexity of setting up Lync, the need for consultants at times, and the fear admins harbor of, say, a fiber line being cut — thus disabling both Internet and phone service. He responded that in the telephony world, it’s not uncommon to have your PBX administration outsourced, especially in smaller companies. Plus, IT typically has to work with a third party for its telephony infrastructure. IT shouldn’t feel as if it must manage all Lync internally and own its full responsibility; instead, it should be open to using outside providers, as telephony folks do.

One of the benefits to working with a third party with both a telephony background and expertise in deploying Lync is that you’re more likely to get a deployment with little change to the dialing pattern. You should mirror the local calling capabilities of the existing system (same extensions and so forth) because the consistency of behavior will help people adjust.

What about this fear of a cut fiber line or a crashed Lync server that takes down critical phone service? Stark said the answer to that risk is a survivable branch appliance (SBA), available from several vendors, which can switch from VoIP calls on the network over to the public (analog) telephone network if needed. It’s basically a backup circuit.

I asked Stark what Microsoft had done to make deployment easier in Lync 2013. He cited two tools: The first is the Lync 2013 Planning Tool, a wizard that asks simple questions like how many users and sites you have. It asks if you are going to use instant messaging, conferencing, and voice capabilities, then provides a parts list of what kinds of servers you need, as well as their high-end configuration needs, external and internal firewall configuration requirements, and the site topology. It also generates an XML file of the high-level architecture that the second tool, the Topology Builder, uses to delve into the nitty-gritty, such as server names, servers, and their roles and locations.

One reason for Lync’s complexity is that it must support different types of deployments and customers. Plus, customers scale differently: Some start with baby steps and deploy just the instance messaging and presence in Lync, then add conferencing. They may slowly decommission PBX systems in a gradual switchover, or they may switch an entire department or a building to Lync. As you scale your rollout, you can use the Topology Builder to help plan the details on the new equipment required by your expanded architecture.

New features in Lync will appeal to admins One of the cool new features in Lync is the ability to provide for higher availability, an area where previous versions fell short. In those prior versions, admins concerned about site failure could make a hard backup, go to another site, and restore it should the first site go down. They could also stretch pools across data centers, but that only worked where you had a sub-30ms connection, which meant the two data centers had to be very close to each other. Lync 2013 allows the front-end pools to share to a paired pool in a distant data center using automatic failover (whose behavior the admins control).

Monitoring in Lync 2013 has also been improved, with real-time monitoring of the servers and every call or IM, with dozens of data points, including the type of device used, the network, and level of voice quality from that connection. Administrators can look across users, network subnets, buildings, and so forth to find areas experiencing audio or video quality issues. In addition, these capabilities were put into System Center so that you can perform synthetic transaction testing to simulate real-time communications activity, both before deployment and after to validate what’s going on in your network and its quality end to end.

Another cool admin feature is the ability to provide a single policy in Exchange to ensure control and compliance across both Exchange and Lync. For example, if an Exchange administrator has to put a mailbox on compliance watch, Lync’s instant messaging is also put on watch.

Although you may need help getting Lync up and running in your environment, I believe it’s worth the effort.

This story, “How to cut through the complexity of Lync 2013,” was originally published at InfoWorld.com. Read more of J. Peter Bruzzese’s Enterprise Windows blogand follow the latest developments in Windows at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter.