Mobile World Congress is always a good time for reflection -- it is one of the few times of year where vendors have an opportunity to pause development, innovation and internal delivery objectives, and unleash their innovation on the industry.
Like critics, other vendors and operators walk the show floor, look at marketing messages, booth designs and demos, while analysts and press scribble notes to help decode trends and industry direction.
As Mobile World Congress approaches, and vendors line up to demonstrate new technology aimed at solving mobile operator and mobile industry challenges, I thought I would take a minute to reflect on what I saw as the largest challenges for mobile operators from a business, technology and operational perspective. The framework is useful for me as I think about how to map Brocade's solutions and technologies to solving business problems, and also useful as I consider the landscape of forward investments that Brocade, and other vendors, should be making.
Maybe you have your own frameworks, or see subtle (or significant) gaps in mine, but, in any event, let's start the dialogue.
Much has been written in the industry about the revenue pressure and margin pressure affecting mobile operators as they simultaneously build out their 4G coverage and watch subscribers shift their service preference towards access-independent, over-the-top (OTT) services. The business challenges for mobile operators are real, whether due to increased competition from those OTT entities, new regulation or market saturation. However, here, I'll point to five revenue challenges that mobile operators, globally, are enduring:
- The shift to OTT services: Most notable and visible is the erosion of SMS revenue due to global, or regional, OTT messaging services.
- Market saturation: This may be more of a developed market phenomenon, and less emerging market, but as mobile penetration reaches or becomes greater than 100%, the subscriber acquisition model is changing -- yielding increasing cost per gross add -- and creative pricing is becoming a leading differentiator. Clear example are zero-rating of particular traffic types (i.e., "Music Freedom" and "Binge On" from T-Mobile US), the selective return of unlimited data plans and competing on total end bill.
- Enterprise purchasing preference: With BYOD policies widely adopted, enterprise IT as a buying center has been virtually eliminated.
- Long-lived relationships: Device subsidies have either been reduced or eliminated in favor of payment plans or outright device purchases, and long-term contracts that kept subscribers linked to a particular network operator are gone. In addition, mobile virtual network operators (MVNOs) are eliminating the one device, one network model.
- Roaming: Roaming revenue was once a cash cow business for mobile operators -- and subscribers had little choice about the roaming networks they connected to. Today, that has changed because the European Union has imposed new regulation (REGULATION (EU) No 531/2012), and alternative roaming providers have introduced a new type of competition in the roaming market.
This list is not meant to sound like a dire warning, though. Remember that the mobile device has become the de facto means of communication. Plus, just about all services previously enabled via other networks -- social networking, payments/commerce, video/television and banking -- are now driven by mobile devices and networks. Without a doubt, at no point in history has the mobile network been more important for enabling our digital lives. And the expected Internet of Things (IoT) has not even arrived yet.
With these services predominantly delivered via mobile networks, though, come a new set of technical challenges in terms of scaling infrastructure -- both the control plane and data plane -- of mobile networks.
First and foremost, it is important to understand the difference between the economics of coverage and the economics of capacity because it is significantly more expensive to provide uniform capacity to mobile devices than it is to provide uniform coverage. Demand for mobile networks is not uniform either in the time or space domain, and ROI for fixed/average demand is faster than ROI for variable demand. As such, mobile networks today rely on capacity planning models that target fixed/average demand, and with the growing reliance on cellular networks and mobile devices, subscribers are left with an uneven service experience.
In addition, there is growing variance in the types of applications and associated traffic mix that traverse mobile networks, each straining the mobile network in different ways. For instance:
- Consumer video requires broadband speeds, while many IoT devices services drive data at speeds slower than dial-up.
- In-vehicle infotainment requires support for mobility at high speeds, while smart meters are stationary.
- Mission-critical services require ultra-reliability, while background data services accept best-effort delivery.
- Voice and video communications require real-time capabilities, while web browsing is largely asynchronous.
As you can see from the above list, today, we are servicing a variety of traffic models simultaneously on the infrastructure -- small packets with delay tolerance, small packets with real-time requirements, large packets at broadband speed, large streaming packets, etc.
When we design networks for broadband, high-speed mobility, reliability, traffic mix or latency, there are specific design considerations for maximizing efficiency and minimizing costs, and these decisions have a polarity to them. Each decision has previously come at the expense of other design principles, such as: control plane scale vs. data plane scale; distribution vs. centralization; and right processor type, transport types and protocols.
Simply put, the problem is not just that there is more aggregate data volume or that there are more devices connecting to the mobile network -- that vastly trivializes the problems in building mobile networks. The real challenge is not the aggregates. It is the variances. The capex investment required to continue to support increasing data volume across an increasing number of devices with increasing variance in service requirements and traffic profiles is the real problem.
— Kevin Shatzkamer, CTO, Mobile Networking, Brocade, special to The New IP