Manage Your Potential SIP Trunking Pitfalls

Still, 32% of respondents have also considered the added features to be an important driver for the business case, and 5% even said that their reasons to embrace SIP were mostly all about the added capabilities.  SIP's support for voice and audio featured prominently in the SIP use-case, followed by desktop video support, messaging, and presence services.  

As with most business decisions, the case for SIP also has some potential downsides that, if managed carefully, should not deter SIP's inertia.  The first issue that may come up is voice quality and overall quality of service when compared to the rock-solid performance delivered by POTS (plain old telephone service.) Because voice traffic is shared with other traffic on an IP network (whether a public or private IP network), network performance factors like congestion and packet loss can degrade voice quality.  

Fortunately, managing voice quality is one of the easiest issues to solve provided the enterprise and service provider networks use appropriate performance monitoring and management tools.  Since the quality of service (QoS) can be affected at many points along the call path, it is important to monitor and manage each possible degradation point.  For example, the session border controller (SBC), responsible for negotiating call set-up, QoS designations, and codec assignments, is one possible failure point. 

In another case, network bandwidth may be overwhelmed or the physical media may be faulty, causing congestion or packet loss.  These potential failure points also exist with traditional POTS trunk lines, but the diverse routing available to IP voice calls adds to the usual management complexity of a traditional time division multiplex (TDM) T1/E1trunk.  The performance tools selected should help mitigate the problems in both the IP network and the physical media layers, including the trunks that connect to the outside world along with the internal network components.  

Another potential downside is how to manage the complexity of a SIP network that manages full-scale UC applications.  The introduction of many new features is a welcome SIP benefit, with the integration of voice, video, and messaging onto a communal network protocol supporting "any device, anywhere, any time."  This complexity increases when multiple vendors are added.  For example, adding Cisco Jabber or Microsoft Skype for Business / Office 365 to a third party VoIP solution adds complexity to network performance and user experiences. 

Complexity is also increased when managing a network that includes legacy POTS and older Voice over IP (VoIP) protocols (like H.323) with SIP.   Many new SIP deployments will continue to carry some voice traffic on a previous generation of technology, so multi-protocol, multi-generational voice and UC solutions are common. Again, performance management tools, session border controllers (SBCs), and gateways are available to overcome this challenge; however, IT managers need to make sure that the performance management tools, SBCs, and gateways provide an integrated solution and that they don't interfere with the critical tasks that other network-control devices need to perform. 

The third key  element that must be managed in a VoIP or UC network is the allocation of server resources. Whether in the cloud or on the premise, VoIP and UC run on server technology, so classic management rules of computing (CPU), memory, and storage apply.  Making server technology just a bit more difficult than the days where purpose-build hardware was the norm, today's virtualization of computing resources can add to management headaches.   

The final potential issue is more a business-planning challenge than a network management concern.  SIP provides the common session control for voice, video, collaboration, and real time communications platforms like WebRTC.  The benefits and concern with SIP trunking are well understood for most; however, how to adopt the future features SIP will support is a bit more uncertain.  Consequently, any SIP deployments and attendant tools used for service management should be as future-proof as possible.  

Because SIP trunks are provided by IP telephony service providers (ITSPs) that are then connected to premise-based VoIP and UC platforms, any approach to managing potential  challenges needs to monitor and, where allowed, manage both sides of the infrastructure.  ITSPs who offer a fully managed SIP trunk or managed UC service need to view beyond the interconnection with the premise-based UC platform.  

IT managers who choose to manage their own SIP trunks and premises equipment will need to also manage SIP trunk session control and performance parameters. If the ITSP doesn't allow for this then, at a minimum, the IT manager should have a management platform that can isolate faults within the SIP trunk.   

IT managers should choose a single integrated management platform that:

In an ideal world, the management platform should also be able to correct faults before they happen.   Management and monitoring systems should also be able to pinpoint the trouble spot, and if automated corrective action isn't possible, the system should alert the responsible work group (e.g. ITSP, Network Operations) to address the issue.  

One thing is sure:  the broad deployment of SIP and SIP trunks is inevitable.  Carriers like AT&T and Verizon are already deploying SIP-based session control as they prepare to retire their legacy networks in favor of VoIP and VoLTE (Voice over LTE.) Most enterprise networks have VoIP deployments in place or have plans to do so.  Third-party cloud providers who offer VoIP, video, collaboration platforms, and WebRTC are also using SIP.  Performance management, interoperability, and security will only become more critical as SIP becomes more prevalent.