Features

June 2008

Telephony

Five tips for successful VoIP deployment

Determining the network's readiness is the first step to migrating to IP telephony.

by Criss Scruggs

Implementing voice over IP (VoIP) in the enterprise requires a solid plan. A successful VoIP deployment requires a careful examination and network assessment, as well as a system to measure and monitor the system in order to benefit from the cost savings, flexibility and functionality VoIP offers.

There are five areas to consider when planning a VoIP deployment:

1. Make sure the network can handle VoIP. When voice communications move from dedicated phone circuits to packetized Internet protocol, voice is forced to compete with everything else on the corporate data network. To prevent conflicts between voice and data, the network should be tuned to prioritize voice and throttle back the bandwidth given to data file transfers-or anything else that does not require real-time transmission.

Determining the network's readiness for VoIP is essential. A readiness assessment to establish the baseline capabilities of the network will help determine which areas of the overall system need to be fine-tuned or upgraded to support the streaming media requirements of VoIP.

2. Keep the deployment simple. Do not deploy a complete VoIP system across the enterprise all at once. Run a pilot program before putting VoIP into production across the entire enterprise, or plan to launch the solution in increments. Pilot programs and incremental rollouts allow companies to determine if the system is working properly and to execute proactive surveys to determine user satisfaction, availability and quality.

3. Create network service maps and update service-level agreements. During predeployment, network administrators should create maps of the network and define service-level agreements with internal departments and external clients. With a proper inventory of the network in the form of a service map, administrators will be able to pinpoint potential bottlenecks and areas of the network that need to be upgraded or extended to support the additional traffic.

Setting requirements in advance is essential, because appropriate expectations and regular feedback between business owners, technology departments and end-users will result in a more successful VoIP experience. Define the policies for ongoing monitoring, performance measurement and management of the network and the VoIP system. Typical SLA metrics include network uptime, application availability, and network and application response time. The data gathered is used to measure pertinent service delivery aspects, such as delay, jitter and uptime, and report confirmation that all requirements and expectations are being met.

4. Consider QoE. Quality of experience (QoE) is a way to understand the user's perception of the quality of the telephony experience, making it an accurate measurement of the success of a VoIP implementation. QoE looks at performance from an end-user perspective to measure how well the network is satisfying end-users' requirements. By paying attention to QoE, the IT department will know about and be able to address users' issues with call quality, intermittent dial tones or technical problems.

5. Review, reassess and repeat. Networks are not static, so implementing an ongoing monitoring process is important. Any change to the infrastructure or usage patterns has an impact to everything on the network. Bringing servers on and offline, upgrading hardware or virtualizing portions of the environment can impact VoIP services.

Continuous monitoring and measurement of IT operations and service-level reporting will provide needed information to quickly resolve network outages and system issues. This technical and business intelligence analysis supports service improvement plans that will sustain the VoIP implementation and make sure end-users do not experience call degradation issues.

Much of the management of the VoIP system and applications can be automated to allow network assessment and monitoring to be repeated consistently to maintain appropriate baselines. Baselines enable administrators to monitor performance and availability, as well as prevent, diagnose and resolve problems.

Criss Scruggs is a senior manager of product marketing at NetIQ, Houston.

For more information (click here)