Business and Finance
Business and Finance
To NFV or Not to NFV? Mitigate Risk, Maximize Profit with Network Virtualization Technologies

As network operators and enterprises assess the level of investments needed to transform their networks, they will contend with uncertainty and risk in many areas. These include traffic growth, pricing levels, regulatory and competitive actions—and above all­—customer demand.


The wildcard of customer demand is a particularly challenging variable as carriers consider new business models in light of shifting customer preferences and reduced barriers to entry in existing markets.  New technologies such as Software Defined Networking (SDN) and Network Function Virtualization (NFV) enable carriers to effectively manage this risk by:


  • Introducing variable costs into their service architectures, such that if demand goes away so does the costs.
  • Aligning investments closer, perhaps to within minutes, of demand such that the certainty of the demand is known, and minimizing network over or under-provisioning.

In the end, what this amounts to is strategic flexibility. Carriers and enterprises can build flexibility into their planning processes such that they are better prepared to manage changes in the external environment.


Make Operating Leverage Work to Your Benefit


The concept that measures or approximates business risk is the managerial accounting concept of operating leverage. Simply stated, the more leveraged a company’s operations, that is, the higher the fixed operating costs are, the more sensitive operating income is to changes in sales levels and pricing.


Operating leverage is the ratio of the fixed costs relative to total costs incurred in the production of a service. Businesses that have a relatively high level of fixed costs, have a higher bar to profitability. By replacing some portion of their fixed costs with variable costs, they can reduce the downside risk associated with lower demand or competitive actions.


To demonstrate this concept, we can plot total cost versus demand for two solutions – one a software-based NFV solution and the other a dedicated hardware platform. This simple example focuses on a range of demand in which fixed HW costs remain constant while variable costs (NFV) rise in direct proportion to demand. In reality, both the hardware line and the NFV line will have a combination of both fixed and variable costs. Our intent is to illustrate the behaviors of variable and fixed costs over a similar range of demand for a network function.


 Figure 1 Operating Leverage Blog.jpg

Figure 1. Fixed vs. Variable Costs


The rise of SDN and NFV technologies gives operators the new freedom to choose the level of operating leverage based on expected demand for certain use cases. The chart shows two zones of demand. Zone A is where you get clear savings from a variable-based NFV implementation, and zone B is where you now have options for continuing with the usage-based model or deploying hardware.  


The theoretical breakeven point divides the zones. This point exists where the total costs lines of a variable cost implementation, say a virtual router on an x86 service platform, and the cost of that network function on a dedicated hardware platform are essentially the same.


If demand for a network function falls consistently within zone A, then you will want to stay with a variable-based cost structure. This case is the ideal solution for faster time to revenue, lower traffic volume use cases, such as virtual CPE, and small and branch office locations.


With increasing demand, you start to enter zone B where you now realistically have other options regarding your cost structure. Once you have passed the theoretical breakeven point, you have the viable option to either switch to a dedicated platform or can continue using VNFs to scale your service.


So why would a carrier or an enterprise consider remaining with an NFV solution when demand is the zone B? Your choice of platform depends on your outlook for future demand growth and maximizing the efficiency of your capital. There are at least two primary scenarios where this could be the case.


The first occurs, as shown in figure 2 when the carrier is simply uncertain about the sustainability of future customer demand. This situation is a common business risk scenario where there is concern regarding competitive actions and rapidly shifting customer preferences. If the carrier overprovisions, then capacity and capital is stranded; if under-provisioned, there is the potential loss of revenue and of not meeting SLA and QoS commitments.


Figure 2 Operating Leverage Blog.jpg

Figure 2. Uncertain Demand


The second is due primarily to the diurnal cycle that occurs in every business, which the figure below illustrates. This pattern, which demonstrates an inefficient use of capital, can also occur seasonally, and is caused by demand that is cyclical in nature.   Another case is where demand is spiked, sporadic or unpredictable. If the demand is in zone B at peak load, then you want to have the options to add incremental capacity on demand and fix function capacity at the lower limit to optimize the use of capital.


Figure 3 Operating Leverage Blog.jpg

Figure 3. Diurnal Demand Cycle


The demand patterns discussed above often apply in the following use cases:


  • vCPE for branch office expansion or managed service offerings
  • Provider Edge where you have variable traffic and demand profiles across a wide base of users
  • Data Center build-outs where the cost of adding additional physical equipment can be high
  • Mobile Core where there is a clear benefit to reduce the cost of under-utilized assets during off-peak hours

In the end, for each of these use cases you have real options to help business decision makers better navigate uncertainty and increase capital efficiency. Like financial options, real options themselves have inherent value.


Fine-tune Your Operating Leverage with Juniper Solutions


While the above analysis provides a simplified view into the business economics to aid managerial decision-making, it is the ability to virtualize networks and automate business processes that enable these benefits.


Juniper built its reputation on performance and scalability, with generations of solutions that scale and increase capital efficiency. Now with our success developing SDN and NFV technologies, we enable network operators to build new services and networks that take advantage of the economic benefits these technologies offer because we have built our virtualized network functions on the same carrier-grade platform as our dedicated platforms.  


To learn more about the economics and assessment methodologies of SDN and NFV, please view our white paper, “Building Your Business Case for Network Virtualization”.