Definition In legacy communications environments, voice and data services operate independently of one another. Voice services are delivered via dedicated circuit-switched networks, while data services are provisioned over their own packet-based networks. Because IT must oversee two sets of infrastructure, capital and operating expenditures rise thanks to hardware and software acquisition and maintenance, along with the need for specialists to monitor and manage the dual enterprise footprint.
The introduction and subsequent maturation of the Internet protocol (IP) is changing the landscape. IP offers the opportunity for all forms of information – data, video, and voice – to be encapsulated within packets. With voice over IP (VoIP), deployment of a next-generation communications solution becomes feasible, where voice and data services co-exist on a single network.
Applications The benefits of opting for VoIP are clear. The converged infrastructure offers the promise of fewer resources to purchase, a more homogeneous environment that is easier to maintain, and more straightforward network monitoring and management, where improved performance and faster mean-time-to-repair result in a more satisfying experience for the user community.
But VoIP has its drawbacks as well. Packets are subject to latency, jitter, and even loss as they traverse the network from their source to their destination location. For data-oriented services, the ramifications of these events are typically minor and often go unnoticed. For voice services, the impacts to call quality can be devastating, rendering conversations unintelligible and putting the investment in convergence at risk.
Implementing VoIP means recognizing that although voice and data can share the same network, they must be treated as distinct entities, giving rise to the creation of different classes of service. Effective VoIP management requires assigning priority to voice packets to preserve call integrity, while simultaneously ensuring data applications continue to perform at levels users expect and demand.
In other words, moving to VoIP isn’t a trivial endeavor, but yields great rewards when executed properly. Therefore, it is imperative to identify critical considerations and follow best practices throughout the VoIP deployment lifecycle.
Key ConsiderationsFirst, ensure that the infrastructure is VoIP-ready by performing a complete assessment of the network, from LANs and WANs to desktops and phones. Baseline and monitor actual usage and performance to determine if the data network must be modified to handle voice calls. During implementation, verify that all elements are communicating properly, from jacks, cables, and phones to switches and servers. And establish a performance baseline by generating synthetic VoIP traffic to confirm call quality prior to full deployment.
After deployment, be prepared to actively monitor VoIP performance from various points on the network, including the core, individual routers, and WAN links. Because VoIP problems can have a myriad of causes – from physical problems on the local loop and overutilized ports to misconfigured class of service settings and high levels of jitter within the voice application itself – institute a broad management and troubleshooting strategy for the converged network. Real-time and back-in-time perspectives are vital, to proactively circumvent emerging situations, resolve ongoing anomalies as rapidly as possible, and identify and correct intermittent problems.
Finally, make the most of the VoIP investment by optimizing the converged network deployment. To do so requires capacity planning and traffic management, performance baselining, and a commitment to continuous improvement. Examining management information from the edge to the core can help make decisions such as if and where to increase bandwidth to account for additional usage caused by VoIP, which applications such as file sharing and streaming media are impeding performance, and how traffic should be shaped so the most business-critical and delay-sensitive services and applications receive priority.
The key to success is selecting a robust VoIP management solution that allows for implementation of all of these best practices activities. If the solution has the breadth and depth to deliver the following, the benefits of VoIP can be maximized and the complexities mitigated:
- Planning for implementation (includes configuration; design; asset management; and call simulation
- Managing operations (includes handling dropped calls; measuring packet loss, jitter, latency, and mean opinion scores of VoIP call quality; measuring quality of service data; and creating policy-based data
- Troubleshooting anomalies (includes capturing all traffic flows and measuring performance across all seven network layers; managing alerts; drilling down and/or back-in-time to isolate root cause; and automating problem resolution)
- Verifying service provider performance (includes service level agreement measurement and tracking
- Monitoring and measuring quality of experience (includes end-to-end visibility of VoIP network performance)
- Interpreting reports (includes analysis of call patterns; visibility into traffic; application layer reporting; and business impact of outages)