Over the top (OTT) voice application providers continue to challenge the traditional voice carriers making it critical for them to differentiate their offerings by delivering high quality and reliable services. An important consideration to support this differentiation that both IP Exchange (IPX) and Next Generation Network (NGN) providers should consider is implementing a true geographical redundant solution for their Class 4 switching network.
In an earlier blog, we wrote about how carriers are deploying next generation networks in a way that allows them to deliver high definition (HD) voice that rivals OTT providers. Yet, it’s not only the quality of the actual voice that’s important, but also availability and reliability of the overall offering that’s critical in delivering a differentiated service. There are many factors that can contribute to a data center, equipment site or network resource being unavailable, from simple fiber cuts or equipment failures to fires and natural disasters. Providing local high-availability solutions are not enough. Carriers need to ensure the network, services, and management capabilities of their VoIP switch architecture survive in the event of a localized failure.
While there are several failure scenarios to consider in developing a sound network and service availability strategy for your class 4 switching network, operators, especially those deploying IMS/VoLTE networks, should concentrate on making sure that the following five critical attributes are present in their network architecture to have a true geographical redundant solution:
- Distributed routing policy: Switching, routing and policy need to be distributed around the network, so that if one site goes down, other sites have the capability to properly route calls across the network.
- Standalone operation: In the event that a site gets isolated from the rest of the network, the site should continue to support local switching until the network is restored. This can only be achieved with architecture that distributes the intelligence across the network.
- Centralized management: Management of a distributed architecture can be very complex. Centralized management of routing and policy is critical to simplify operations and ensure consistent routing and policies are enforced
- Distributed switching fabric: Call control elements need to support 99.999% availability, even when geographically distributed – that translates into only a little over 5 minutes per year downtime. This kind of high availability especially with catastrophic failures affecting an entire site can only be supported with a distributed switch fabric
- Redundant components: Non call control elements, such as element management systems, event collectors and call detail record collectors, are not involved in the actual call setup and are not needed to maintain the calls, but they support crucial functions and contain critical information. Geographical redundant solutions should also be provided for these components to allow for quick recovery of failures.
We have all experienced dropped calls, low voice quality and periods of unavailability using the popular OTT applications, but carriers incorporating the above attributes in their networks have the advantage of being able to provide higher valued services compared to those from OTT competitive threats. Are you experiencing the benefits of true geographical redundancy in your network? Tell us what you think, and suggest other attributes that need to be considered. Tweet them to us at @Dialogic. You can also check out a video on how carriers can deploy true geographical redundancy.