 | Issue: | EMEA 2014 |
| Article no.: | 3 | |
| Topic: | NFV – Enabling accelerated innovation | |
| Author: | Niall Norton | |
| Title: | CEO | |
| Organisation: | Openet | |
| PDF size: | 172KB |
About author
Niall Norton, CEO, Openet
Niall Norton joined us in March 2004. Mr. Norton has served as our Chief Executive Officer since September 2006 and has served on our Board since August 2006. Between 2004 and 2006, Mr. Norton served as our Chief Financial Officer.
Prior to joining us, he served as Chief Financial Officer and Corporate Secretary of O2 Ireland, the Irish operations of O2 Group Telecommunications PLC, from 2001 to 2004.
Mr. Norton holds a degree in Commerce from University College Dublin and is a Fellow of the Institute of Chartered Accountants in Ireland.
Article abstract
The benefits of investing in next-generation policy management are clear, with more efficient use of network resources and the ability to create more innovative, personalised services. However, the cost and complexity of implementing these solutions can deter some operators, particularly in markets where CAPEX spending is limited. One solution is a hosted deployment model, which offers a faster and more efficient way to bring services to market while providing a more predictable cost structure. This is an appealing proposition to those operators that lack the resources or the budget to embark on a more traditional deployment of policy control and enforcement solutions.
Full Article
Mobile operators of all sizes in all parts of the world have a lot in common. They’re seeing increased competition from OTT (over the top) providers, they need to increase customer revenues and margins and increasingly they are running lean operator models. They need to quickly turn innovation into revenue.
In many of Europe’s mature mobile markets that are characterised by (relatively) high ARPU, data is now the main differentiator as voice and texting become commoditised. It’s in the area of data that operators need to quickly develop innovative offers with flexible pricing and move towards providing a better customer experience, through real-time context aware offers and communication. These can range from a simple, ‘you’ve almost reached your data limit, would you like to buy an additional 500MB for €5’ offer, to a much more complex offer upselling of personalised content services triggered by an event.
Declining smartphone prices are leading to increased data usage in all markets, even those without a large percentage of high ARPU users. We’re already seeing operators sell low cost data offers promoting OTT services such as WhatsApp to encourage data adoption. This increased data usage driven by lower cost smartphones and tablets and increased network speeds is seen by operators as the next big opportunity.
Successfully catching the data wave means having the ability to launch more services in less time. It means being able to quickly explore new offers for such things as wearable technology, connected cars and devices, and digital lifestyle services. It means being able to work with and also compete with OTTs, and rapidly develop new business models and partnerships such as sponsored data, and it means a real-time understanding of customer activity so that context sensitive offers can be made. It’s a long way from selling voice minutes and text messages. In order to do this, operators need to innovate and their ability to quickly provide new innovative services is tied to the agility of their BSS (business support systems). This includes functions such as charging, policy management and billing. In some cases buying, implementing and managing these systems was relatively expensive and the most advanced systems could be out of reach of some operators in low ARPU markets. Even in the high ARPU markets with a high percentage of existing data users, the ever increasing volume of data usage is forcing operators to accelerate innovation and devise new ways to differentiate their offers, provide better value to their customers and increase data revenues, as well as keeping up with the scale requirements that increased usage is placing on BSS.
These challenges are causing difficulties for operators who need BSS to enable increased levels of innovation in an ever decreasing time scale. It’s also causing problems for traditional BSS vendors who can take several years to specify, adapt, text and implement their systems, as the functions specified could be out of date by the time they go live and woefully undersized for current data volume requirements.
This is where Network Functions Virtualization (NFV) can help.
Network Functions Virtualization (NFV)
Network functions Virtualization or NFV was originally conceived by the world’s biggest operators to enable reductions in both capital expenditure ad operating cost. The cost savings are achieved by moving communications technology away from proprietary systems towards open systems using commercial off the shelf (COTS) hardware. In addition, the aim was to improve flexibility and reduce the time to roll out services by deploying in a virtualised environment. Most of the traditional suppliers to operators (and some innovative new ones) have now embraced the NFV concept and claim to offer NFV compliant solutions. The NFV initiative has allowed a variety of new deployment models to be considered.
Private Cloud Deployment Model
Large Tier 1 operators operating in mature markets need to keep up with the scale requirements. However, established deployment methods that use long term forecasts to budget for expansion struggle when faced with the never-ending battle to build out a network that is capable, flexible, and fast enough to meet insatiable customer demand. Indeed, this very challenge prompted the operators to launch the NFV concept in the first place.
My own company has worked with a major Tier 1 operator in North America to deploy a fully virtualised policy management solution. While this is very early days for NFV, the scale of this deployment is noteworthy, currently supporting 100m subscribers and thousands of transactions per second (TPS) in real time. Here, policy control is delivered as software-based virtual network functions that are combined in modular scaling units. An operator taking this route doesn’t have to predict today what it is going to need in several years. Instead the immediate requirement can be established, and computing resources added to the private cloud infrastructure in the data centre by simply adding commodity “off the shelf” Intel x86 based servers.
The orchestrated deployment of virtual network functions utilising discovery and auto-install capabilities, removes the need to match software components to specific physical servers. Therefore, the virtual policy deployment also eliminates the need for verifying operating systems, patches, and components in a pre-deployment test scenario that can add months to the deployment timeframe.
Hosted Deployment Model
The benefits of investing in next-generation policy management are clear, with more efficient use of network resources and the ability to create more innovative, personalised services. However, the cost and complexity of implementing these solutions can deter some operators, particularly in markets where CAPEX spending is limited. One solution is a hosted deployment model, which offers a faster and more efficient way to bring services to market while providing a more predictable cost structure. This is an appealing proposition to those operators that lack the resources or the budget to embark on a more traditional deployment of policy control and enforcement solutions.
The Hosted Deployment Model utilises the same concept as a dedicated “Private Cloud” infrastructure in that ‘Network Functions’ are deployed as virtual applications but with the infrastructure being hosted by a third party partner and consumed like a cloud based SaaS (Software as a Service) model. This brings new challenges, of course. For example, hosted Network Functions may require real time responses that are difficult to achieve when the network functions are geographically dispersed. However, this is solvable as most operators will already have arrangement with partners such as Internet Exchange Carriers that provide transport services for international interconnections. These IXCs have the physical infrastructure for hosting via their Point of Presence facilities and the high speed data pipes to the carrier. Certainly, ‘policy as a service offerings’ via the cloud can be offered via a pre-integrated policy management package delivered flexibly and cost-effectively via the cloud. Deploying using NFV concepts provides considerable advantages to both the hosting company and operators. For example, virtual application deployments allow the hosting infrastructure to be truly multi-tenant as any one tenant can be deployed and managed as a logical entity completely independent of the physical infrastructure and completely invisible to other tenants. There is no longer a need to permanently dedicate physical resources to any specific operator, as now resources can now be easily reallocated to operators as demand requires. This also allows new flexible license options to be considered, such as a “burstable” license model to handle periodic peak demand.
Appliance Model
A powerful advantage of virtualisation is the ability to deliver networks functions as software modules. This enable a Network Function to be simply deployed on a Standard Business Server. Offering solutions like policy as an appliance can therefore help meet needs of small or medium operators. Thus, an operator with only a few thousand customers can efficiently deploy the same technology as an operator with 100 Million subscribers. It is only the degree of scale, in essence the number of servers that differs. A good example is Tele Greenland, the only telecoms provider in the least densely populated country in the world (population c. 56,000), which is using a pre-packaged PCC solution to roll out flexible charging plans tailored to the activities of individual subscribers.
Even the biggest operators can benefit from this approach – e.g. for adjunct or niche business needs such as specialist MVNO or M2M use cases, allowing a fast, cost effective service roll out that allows for easy expansion. Because the applications on the appliance are deployed as virtual Network functions, they can be easily migrated in to a larger Private Cloud infrastructure.
Conclusion
Network Functions Virtualisation can deliver benefits irrespective of the deployment model – via a substantial Private cloud infrastructure, a hosted infrastructure or an appliance.
Alongside significant cost savings, NFV can bring faster time to market and reduced risk when deploying new services: These can be rolled out or rolled back, or easily upgraded without committing valuable resources.
NVF can also provide resiliency: Reduced need for specialist hardware coupled with the affordability of commodity hardware allows for more redundancy, including more cost-effective geo redundant systems. The ability to quickly migrate or roll out (or roll back) additional virtual functions improves the ability of an operator to respond to failures, maintenance issues or unexpected traffic loads.
NFV enables all sizes of operators to have the functionally rich and agile BSS that is needed to drive innovation and therefore maximise the opportunities presented by the explosion of data traffic worldwide.
