Network outages happen! A Heavy Reading survey of mobile operators found that they are spending significant dollars – over $15 billion a year - addressing service and network outages. The same survey identified that the three major causes of network outages were link failures like cable cuts, network overload and network equipment failures with the last cause resulting in the most severe outages.
Take for example the Asia-America Gateway (AAG) submarine cable. It connects the United States with multiple locations in Asia and is owned by a consortium of 19 carriers. Last year alone, it experienced three service impacting cuts, and already this year there was another incident that the operators initially estimated would take 20 days to get the cable fully back in service. I myself can attest to the devastating effect of what we affectionately called “backhoe fade” on network communications having had the responsibility for a group of experienced technicians that had to make repairs to long-haul fiber cables. Backhoes are the not only culprits; there are other causes, some highly bizarre (as shared in this Level 3 blog), that result in damaged or cut cables. The point is that regardless of whether connectivity is through land-based or submarine-based cables; there is the possibility for outages that can be large in scope and long in duration.
Many customers have negotiated Service Level Agreements (SLAs) with their carriers that provide payback in the event of recurring outages. In a competitive telecommunications environment carriers are agreeing to these types of contracts to win and retain their customers. Some enterprises have been able to require their long distance carriers, in the event of large scale outages for durations lasting from only 15 minutes up to a few hours, to provide compensation in the amount of the entire monthly charge the customer would pay for the affected end points and additional charges for each end point out of service!
Churn is a multi-million dollar issue for the large operators. For Deutsche Telekom’s German mobile operations, the monthly churn for its contract customers in Q3 of 2014 was 1.6 percent on a base of about 22.8M subscribers. Their reported cost to acquire each contract subscriber was about $150. Do the math and you can easily comprehend the enormity of money spent to make sure its overall subscriber base did not decline. And it even had the lowest churn of any carrier in Germany. We’re talking millions of dollars being spent by each of the major carriers each month on churn. While the causes for this problem vary, poor Quality of Service (QoS) and low reliability have a large impact on a customer’s propensity to churn.
It’s apparent when you look at churn and SLA requirements that a carrier’s strategy to differentiate on quality to retain customers and reduce OPEX is only part of the picture; they also need to be able to provide assurances for reliability and availability to even win customers in the first place. Therefore, it’s important that a carrier’s underlying architecture provides the resiliency against network outages both from cable cuts and equipment failures in order for them to deliver high QoS, high availability as well as meet and exceed the performance requirements they may write in their SLAs.
At Dialogic, we’ve incorporated something called true geographical redundancy in our IMS MGCF next generation softswitch platform for those very reasons. It has a distributed switching architecture unique to the market that allows us to provide service continuity for an operator’s network in the event of a catastrophic failure either at the link level or with a particular node. A carrier Implementing a network with this capability can maintain global uptime in standalone mode if a site gets isolated, or continue operations if a network failure event occurs at any given site. Is your network able to do this? Check out this video on how to “Differentiate Your Services with True Geographic Redundancy” and tell us what you think. You can tweet us at @Dialogic.