• Core network enhancements can benefit enterprises from a network performance and service cost point of view.
• Flexible bandwidth services and pricing models are maturing and are worth a second look.
In 2015 and early 2016, SDN was the buzzword du jour of the telecoms industry, but the attention has now shifted to SD-WAN. Perhaps this is inevitable since SD-WAN is the newer technology and is at the forefront of several recent or upcoming service launches from providers such as Telstra, Orange Business Services, and BT, amongst others. SD-WAN also seems to offer more tangible benefits to the average enterprise customer, particularly those with a large number of smaller sites, or those seeking to adopt virtualised network functions such as firewalls and session boarder controllers.
SDN on the other hand has tended to focus on the core network. Talk of separating the control plane and the data plane of enhanced orchestration doesn’t seem to offer much direct advantage to the enterprise. Indeed, many providers have shied away from talking about SDN as it does not easily fit in with their increasing emphasis on offering business solutions rather than technologies. More than this, providers have focused on SDN as a way of delivering internal efficiencies through enhanced automation with their networks, rather than SDN as the basis of new product/service launches.
Enterprises, should, however be asking providers about their SDN capabilities. First and foremost, increased efficiency in the core network will deliver enhanced performance. Reducing costs whilst maintaining margins also allows providers to compete more keenly on price whilst maintaining investment in their networks. SDN’s position at the heart of a provider’s network points to the ability of that network to deliver agility when it comes to provisioning new circuits/connections and on adjusting bandwidths.
Flexible bandwidth has enjoyed mixed reviews. Much was made of the ability of SDN to deliver network-on-demand, a full pay-as-you-go model. The hype, as is so often the case, has been ahead of the reality; but the reality is moving close. Colt, for example, is one provider that continues to push as flexible bandwidth model. Colt is helped by the fact that it has focused on connecting its network into a large number of business premises. Colt states that it is now connected into 24,500 buildings, including 700 data centres globally. This points to the heart of one of the challenges for flexible bandwidth and instant provisioning of connectivity – for it to work, the equipment already has to be in place.
The second problem has been pricing models. Truly flexible bandwidth has not taken off as those providers who offer it often tag on a significant premium (maybe 20%) for a pure pay-per-use service. The model that is now emerging mirrors the voice-minute market in that customers are able to pay for a fixed bandwidth service (e.g., 100 Mbps) and then flex that up (e.g., to 200 Mbps) as needed. This can be charged for on an hourly rate and controlled via a user portal. It can also be programmed to flex automatically at known peak times (e.g., end of month reporting, or scheduled data backup). The real value may come from application-driven bandwidth flexing, but for now SDN-enabled flexible bandwidth is worth a second look.