The success of the cloud has been lauded by the enterprise sector for quite some time. Cloud technology has disrupted this market with its cost and scale efficiencies. The technology enables a rapidly deployed, shared pool of hardware resources to be leveraged. This delivers the ability to scale up or down and facilitate heterogeneous workload consolidation; all while driving hardware utilization up. Operators now face an increasingly challenging market in which competition is intensifying and financial fluency is constricting. It comes as no surprise that carriers are looking to the cloud model for their own central offices. However, constructing a cloud architecture that caters to the demanding needs of the telecoms space is not straightforward.
Central office optimisation
The roadmap for the telecom cloud is well and truly under way. Leading operators around the world including AT&T, Verizon and Deutsche Telecom, have laid down their own vision for the cloud in the Network Function Virtualization (NFV) white paper that was released in October of last year. This NFV vision is now being carried forward by the European Telecommunications Standards Institute (ETSI) Industry Specification Group (ISG). Many operators are already building their cloud to enable them to provide services, such as IaaS and PaaS, to the enterprise sector. Operators are uniquely placed to do this – as they have the connections with CIOs and already deliver broadband connectivity services to enterprises. However, the advent of the cloud in telecoms also now enables operators to deliver combined connectivity and computing services to enterprises.
But this cloud opportunity goes much deeper for operators than enabling them to provide an increased range of enterprise services. The cloud unlocks the opportunity to streamline the central office, releasing pressure on CapEx and OpEX, and helping operators become sleeker in their operating practices and compete more efficiently in a crowded market. Cloud technology offers a path to a completely new, scalable and cost effective kind of telecom infrastructure through decoupling hardware from software. The central office of tomorrow will not deliver network functions as integrated boxes. Instead, new network functions will be deployed as software workloads on homogenised hardware platforms. This amalgamation of network function workload will drive CapEx down and utilisation up. This will also positively impact the rate at which operators can launch new services; but also enable them to cost-effectively trial new services while still ensuring they are prepared to scale any service, should it become popular, through the cloud’s elastic architecture.
Only certain parts of the network are best suited to virtualization and workload amalgamation.
Where to leverage Virtualisation
Only certain parts of the network are best suited to virtualization and workload amalgamation. For example, there are functions that construct the pipe, like SGW, PGW, SGSN, PCEF etc. Data plane functions like this manage high packet throughput and often require low latencies. Next are functions that control the pipe such as softswitches, SBC, CSCF, PCRF etc. These control plane functions manage signaling plane traffic and often need ‘five nines’ high availability. The third area is application servers that provide specific services over the pipe. Everything in the service plane, as well as BSS/OSS nodes, lends itself very well to workload consolidation.
‘Five nines’ high availability is just one of the crucial requirements the telecom space has, alongside high reliability, low latency and provision for regulatory requirements. Not forgetting specific needs around timing and synchronisation. These factors are critical to the smooth operation of the wireless network. While the industry should champion the principles of the cloud – it must still ensure that the telecom cloud provides all the requirements needed for successful network operation. The public cloud leveraged by so many other industries is woefully ill-equipped to stand up to the high requirements of telecoms from a central office point of view. The superior choice to enable these functions is a private telecom cloud which virtualizes network functions.
ATCA – The Enabling Platform
Once an operator has identified the network functions it intends to virtualise, and the workloads it would like to consolidate, it then faces the challenging task of how to construct its private telecom cloud. An operator must select a best-of-breed platform to build its cloud that provides all the central office requirements of the modern carrier we previously mentioned. As well as delivering these factors, the platform an operator chooses must also provide the highest level of scalability, reliability, virtualisation and workload consolidation and cost effective elasticity.
The most suitable platform to ensure this is delivered is ATCA. An ATCA platform couples telecom infrastructure requirements with virtualisation and load-balancing to deliver telecom workload consolidation in the central office. Its protocol aware telecom load balancers allow operators to consolidate heterogeneous workloads like PCRF, HSS, MME, SBC, CSCF, DRA etc, on one homogenised ATCA platform.
This ‘best of both worlds’ approach in this type of platform can ensure operators are able to rapidly realise their NFV vision in their central office, without compromising on any strategic telecom infrastructure requirement, and become leaner and better competitors.