article page | 1 | 2 | 3

applications, many of which may be unknown and unauthorized.

Using deep packet inspection or NetFlow technologies, network managers can create an accurate inventory of applications and establish the capability to identify and proactively monitor business-critical applications. Doing so enables a network manager to understand which applications run on a network, traffic trends and patterns, where, when, and why latency occurs, and the quality of service (QoS) metrics associated with an application. The tools used must not only have the capability to identify applications and traffic patterns, but also classify applications by priority and criticality.

With a more-accurate picture of how the network is used, IT managers can then optimize the infrastructure prior to installing new services or applications. This might entail curtailing rogue activity or unauthorized network usage, adjusting traffic prioritization through router settings, or other QoS control mechanisms.

From Pilot to Production

The initial analysis and monitoring phase also provides a network baseline that will serve as a critical frame of reference once voice or video traffic is introduced. Network managers can use this baseline to ensure the quality of pre-existing applications throughout the change process. Because VoIP creates a lot of traffic and can be more fragile than other services, network managers should put in place purpose-built monitoring tools that collect granular metrics needed to assure the quality of convergence technologies. In the case of VoIP, this means the ability to capture all call records, associated mean opinion score (MOS), and other voice-specific metrics that enable analysis of individual call problems in the context of overall bandwidth usage and call volume.

In Summary

Best-of-breed monitoring tools should be put into place to enable the setting of performance thresholds so that most problems can be averted before end users call the help desk. These tools will accelerate problem resolution when issues arise, and provide engineers with the data to plan capacity as the use of voice and video grows. Companies that have postponed managing the applications in their networks should begin now to evaluate flow-monitoring technologies as a low-cost way to protect their application investment and ensure a positive end-user experience.