How to Handle SIP Trunk Failovers

As the growth of VoIP and SIP trunks continues to increase, businesses are exploring the benefits and disadvantages that this new technology offers. SIP trunking provides flexibility, scalability and costs much less than traditional enterprise phone systems but it is also a business application like any other. It means that businesses must consider aspects like disaster recovery, resiliency and business continuity with regard to SIP trunks.

SIP Trunks versus Hosted VoIP

It is relatively easier to design backup and recovery plans for hosted VoIP since most of the intelligence lives in the cloud and the vendor maintains the infrastructure. Specifically for small and medium businesses, it is normally sufficient to maintain an alternate Internet connection with a different ISP vendor to ensure continuity.

SIP trunks on the other hand can range from simple to complex networks. Even with SIP trunks, small businesses can usually manage with using mobile connections as a backup option but this is rarely feasible for any organization with more than a handful of employees, multiple locations or more complex routing requirements.

Handling SIP Trunk Failovers

With any business process, designing backup plans is something that the company should consider well in advance of implementation. Unfortunately many enterprises consider backup only after the system has been built/purchased and deployed. By then it is too late to modify the system if needed and the business will have to compromise in some areas of recovery. If the first time your business thinks of backup is during an emergency, naturally it is too late!

When considering any failover plan for SIP trunks, managers need to keep 3 things in mind –

• The cost of providing effective backup, continuity and recovery services against the risks of service disruption
• Define the conditions under which an event is classified as a service failure (ex: should we activate the backup if phones fail for 2 minutes? Or 20?)
• What is the minimum service level required so that the business can continue functioning with the most critical operations?

Mapping Call Flows through the SIP Network

There are many components involved in routing calls between users and any business continuity plan should account for failure in one or more of these components – whether or not there is an actual disaster. For instance, hardware failure can occur at any time, not just if there is a fire or flood in the building. Components such as IP servers, session border controllers, the enterprise LAN/WAN, routers, IP phones etc. can fail, causing a service disruption.

In order to ensure that the SIP trunks failover gracefully if something fails, we need to map call flows through the system. We can do this with software or even on pen and paper, depending on the complexity of the network. We need to map all kinds of calls – between employees at the same location, between users at multiple locations, calls coming into the enterprise and going out and so on. Once this exercise is complete, we will be able to visualize and understand what kind of features we need for the backup system.

What Are My Options for Handling SIP Trunk Failovers?

The availability of failover options varies between providers and not every vendor will provide comprehensive failover protection in case of service failure. Some of the common options available in the market include:

Forward calls to another location

If the primary SIP trunk group for a business fails, some vendors offer automatic or manual forwarding to an alternate destination. The alternate destination may be a mobile number or another SIP trunk.

Separate trunk groups for critical services

Your company may choose to purchase two separate groups – one for general use by all business departments and one for critical services (usually sales, customer service etc.). This solution works great for those situations where departments occasionally have to deal with high call volume. Even if one group is completely utilized, the other departments can continue functioning.

Load balancing

Some vendors offer integrated SIP trunk groups with load balancing features. It means that the service can allocate calls between multiple SIP trunks to ensure that no single trunk reaches hundred percent utilization. Therefore load-balancing is a good option even during normal business hours to ensure consistent utilization across all SIP trunk groups.

Failover to PRI lines

This is often considered to be a comprehensive solution but it is also complex and expensive. If the SIP network fails for any reason, calls are automatically redirected over standby PRI lines which preserve caller ID and other information as well.

We can see from the above options that each has its own benefits and costs associated with it. Any business wants the best continuity options but the costs may be too high to justify. Hence weighing the costs against the risk of complete service disruption is an important part of handling SIP Trunk failovers.