This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter's approach.
Converged business communications environments are increasingly comprised of a complex mix of VoIP, unified communications and SIP trunking technologies from a variety of suppliers. And whatever that mix is today, it is likely to change going forward.
As a result, IT teams are faced with the daunting task of ensuring all these diverse moving parts work together to deliver the necessary voice quality and service levels -- without breaking the bank in the process.
So, how do you reduce the complexity of these business critical communications environments and cost effectively manage their operation? And when voice quality and service levels do degrade, how do you quickly and efficiently answer the two most important questions: where and why?
To answer these questions, operation teams require two key weapons in their arsenal -- visibility and metrics.
Converged communications blur the traditional lines of voice and data, which historically were supported by different groups within the organization. Today, this approach no longer works, largely because VoIP and UC are highly dependent on the underlying network they traverse, and voice quality and service issues are often caused by a variety of network-related conditions.
When these problems occur, support teams that must gather, analyze and compare data from multiple sources and locations across an assortment of routers, gateways, VoIP call servers and UC applications, are at a serious disadvantage. This patchwork of disjointed information magnifies the complexities of the environment, complicates support efforts and increases overall operating costs -- not to mention that it diminishes responsiveness and the ability to quickly resolve communication problems.
Support teams today need end-to-end visibility -- across every facet of the communications environment, from the SIP trunks at the core to the softphones at the edge -- in a consolidated, unified view.
Through unified visibility, support teams can determine the effects components have on each other, and more accurately isolate "where" quality and service issues originate -- whether it's at the SIP trunks before traffic entered the network, or within the core UC components such as the IP-PBXs, messaging applications, call centers or endpoints.
The ability to see into the entire converged ecosystem and understand the relationships between the components and applications is a step toward not only reducing the overall complexity of the environment, but also reducing the operational costs associated with delivering communications service levels that support the business.
Use voice-specific metrics
When the network becomes responsible for moving voice and advanced communications payloads, the traditional SNMP-based approach to supporting and managing the network is no longer sufficient.
Voice traffic subjected to common network anomalies, such as latency, delay and packet loss, can result in dismal voice quality, impair conversations and frustrate users. It doesn't take long for these conditions to drag down productivity and increase overhead costs. As operations teams tackle the support load of the VoIP and UC core network, the need for voice-specific metrics that provide clear insight into "why" quality and service issues are occurring becomes crucial.
A common approach is to rely on network mean opinion scores (MOS) and quality of service (QoS) metrics averaged over time. But these aggregated averages can actually hide quality problems in a real-world deployment, causing the support team to waste time searching for the true cause of the problem, or potentially making changes that don't resolve the real issue.
For example, a 15-minute phone call might result in a decent or acceptable MOS score. But that call may have experienced a few intervals of burst, which are times of excessive loss resulting in severe quality degradation. During those bursts, the user may have missed 20-30 seconds of the conversation. Now if that happens at a significant point in the conversation, the user's experience would not align with the overall MOS score and the support team would be at a disadvantage to uncover why the user experienced poor quality.
When users complain about a specific call, support teams need deep VoIP-specific metrics, such as burst, length and density on a per-call basis, to determine the type of voice issue occurring. Once the issue is determined, the team must determine what network conditions or endpoint issues may have contributed to the problem. Only then will the team definitively know both where and why the issue is occurring, and what to change to resolve the problem.
Case in point: Say a call fails over a specific trunk connected to a gateway. Do you know if the channels are up or down? Did you get an alarm on any of those channels? Or perhaps system resources on a particular switch are extraordinarily high. Do you know if you are experiencing any packet loss for VoIP conversations and how bad the problem is?
Through unified visibility of the entire environment and deep voice metrics, the complexity of determining where and why communications issues are occurring is significantly reduced. Armed with complete, correct information, support teams can pinpoint and resolve the real issues much more rapidly and cost effectively.
What's the bottom line? VoIP quality and converged communications service levels are improved, support teams are better able to cope with the complex environment, and overall ROI on convergence investments get a welcome boost.
Tone Software Corporation is a global provider of comprehensive network monitoring and management solutions for converged telecommunications and IT infrastructures.
Read more about lans and routers in Network World's LANs & Routers section.
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.