Virtualized Networks Key for 5G

Mobile networks need to virtualize so service providers can better address a wide range of use cases beyond smartphones, including low-bandwidth machine-to-machine apps, low-latency services as well as ultra-high bandwidth scenarios, said Lynn Comp, Intel senior director.

Without automating, such flexibility cannot happen, at least not with manageable operating costs. Networks will need to spin up, and spin down services automatically, said Comp. That means network functions virtualization (NFV) are essential for 5G success.

And, fundamentally, NFV means disaggregating hardware and software.  

The next three to four years will witness mobile networks evolve in a way that is more transformational than was the shift to IP in mobile networks, ABI Research argues.

According to ABI Research, technology trends that address these challenges today and also lay the groundwork for 5G in future include

• bthe application of virtualization and software-defined networking technologies to mobile networks
• the  evolution of the mobile edge to improve customer quality of experience (QoE)
• the creation of monetization opportunities for the operator
• the growing need for an effective self-organizing network (SON) solution
• the use of big data analytics to leverage granular network data to enhance network performance

Stagnating average revenue per user and increasing network traffic are driving operators to be more cost-effective and innovative in network performance and operations management and network upgrades, said Prayerna Raina, ABI Research senior analyst.

IoT Impact in 3 Years?

It is reasonable to suggest the greatest value from internet of things applications will come from insight gleaned from all the raw data, not the actual creation of data or connections to support the gathering of that data.

A survey of enterprise executives conducted by the Economist Intelligence Unit found 38 percent of respondents believe the internet of things will have “major impact” in most markets and industries within the next three years. Some 40 percent believe IoT will have some impact in a few markets or industries within the next three years.

With the caveat that even industry insiders can be wrong about technology impact, and the further qualification that the impact of IoT within a three year period was the framework, there is significant belief that the insights gleaned from IoT will have impact on some to most industries.

As always is the case, we are most likely to err on the time frame. In most cases, gaining business value takes longer than most expect. On the other hand, important new technologies virtually always have greater impact than initially forecast.

source: Economist

What if 4G Emerges as the Key IoT Network, Not 5G?

Business model viability is perhaps the biggest single challenge most service providers now face, as legacy revenue sources that drive the business mature and decline. In that regard, 5G is  viewed as a strategic platform to support new machine-to-machine, low-latency and other services requiring ultra-high bandwidth.

At the same time, the importance of legacy solutions is hard to avoid. Consider that the bulk of existing internet of things apps run on 2G networks, for example. Of course, most believe the huge wave of growth is yet to come.

Ovum’s forecasts suggest mobile network machine-to-machine (M2M) revenue (not including NB-IoT) will reach 733 million globally by 2021, with global annual revenues of $67 billion in 2021.

Keep in mind that this forecast includes all service provider revenues earned from IoT services, not simply the value of the access connections. One might argue that the key variable is the ability to generate revenue “up the stack,” from platforms, applications and services.

The greatest revenue will be earned in Asia, North America and Western Europe, Ovum predicts. IoT connection revenues will total US$ 22 billion in Asia,  US$ 16 billion in North America and US$14 billion in Western Europe, Ovum predicts.

Of all current mobile generations, LTE will be dominant in the long term, accounting for 212 million connections in 2021, Ovum predicts.

That should provide an opportunity for reflection. If 5G business opportunity is premised in large part on IoT services, that reliance on 4G, if it emerges, will limit the 5G revenue opportunity.

In 2021, 2G and 4G will be at a point of parity, while 3G will account for 172 million connections, Ovum predicts.

.

source: Ovum

SD-WAN a Substitute for MPLS?

As hot a trend as suppliers believe software-defined wide area networks (SD-WANs) might be, there is a recurring question about the impact of new SD-WAN deployments on existing business data networks such as MPLS.

Some argue SD-WAN will have most impact in the smaller business or remote enterprise location segment of the market, in terms of new account growth, in large part because SD-WAN might enable deployment at lower cost.

Others might argue SD-WAN also might limit the growth of new MPLS deployments, as SD-WAN will allow aggregation of location bandwidth (internet and MPLS). So, in some cases, additional capacity requirements could be met by adding SD-WAN, not additional MPLS resources. In that view, SD-WAN will have the primary impact of limiting additional MPLS growth.

Perhaps some will argue that SD-WAN can replace MPLS. As with most new technology trends, all three developments will happen, simultaneously. Longer term, the issue is less clear.

Many important new technologies develop first as lower-functionality alternatives to existing high functionality platforms. Over a period of time, those new alternatives begin to move in the direction of higher quality and more features, until, eventually, they become full substitutes for legacy products.

That is not a certainty for SD-WAN, but it is unwise to ignore human cleverness and platform evolution. Still, most would acknowledge, as a sheer matter of physics, that SD-WAN cannot control core network quality of service, whatever it is able to do at the edge of the core network.

The issue is how many use cases that will be good enough to support.

Value or Price Driving Video Cord Cutting?

Customer account losses in the traditional pay TV ecosystem could soon accelerate to around five million a year in the U.S. market, up from about two million a year, RBC analyst Steven Cahall said. A survey conducted by RBC found that only around 55 percent of respondents planned to keep buying services from their video services provider.

The data suggest that “the floor that ultimately will keep the linear bundle” is only at around 68 million homes. SNL Kagan says current U.S. pay-TV homes number 86 million at the moment.

Precisely why consumers are cancelling subscriptions, or refusing to subscribe in the first place, typically is said to be the result either of diminishing interest in the product or high prices.

Do consumers cancel their entertainment video subscriptions because they are deemed too costly? Some say so. Others might argue that disinterest is the issue. So it might seem as though there are conflicting views about why some consumers are abandoning linear video for streaming alternatives.

Maybe not. Perhaps, as always, consumers simply are evaluating products from a value perspective. When consumers say the price of a product is “too high,” what they often mean is that perceived benefits are too low, in relationship to price.

If so, those who argue price is the main issue, and those who say demand for the product is the issue, could both be correct.

Peak Telecom is Coming

Most observers of global telecom revenue will note that, with a couple of possible exceptions, industry revenue has grown continuously, for as long as we have kept records.

On the other hand, one has to wonder whether telecom revenue will reach a peak at some point in the relatively-near future, as mobile adoption reaches saturation in every country and as every customer buys as much internet access as they prefer.

Looking only at the country of Malaysia, the trends are clear enough: Mobile growth is reaching an absolute peak, as is mobile broadband. Fixed line voice is declining, and has been dropping since about 2000, while fixed network internet access has grown to replace the lost fixed network revenue, but itself is nearly saturated.

source: Researchgate.net/

source: FCC

That does not mean service providers will stop innovating--or trying to do so--or seeking to add big new revenue sources. But that new revenue will mostly balance lost revenues in the core business, as voice, messaging and eventually, even internet access revenues fall.

Indeed, replacing lost revenue now is a major industry challenge. The global telecom industry is about a $1.5 trillion annual revenues industry. To move the needle, any new sources have to be large, simply to replace lost revenues from legacy sources.

This trend is seen in many developed markets, but now also can be predicted for fast-growing Asia Pacific markets as well.

http://slideplayer.com/slide/11399351/

Roughly speaking, to sustain three percent annual revenue growth, and assuming zero losses in all legacy sources, some $45 billion has to be added every year. But that is not realistic. With actual declines in voice and messaging revenue, and coming shrinkage and margin compression in newer sources such as internet access or video entertainment, service providers might have to replace as much as half of all current revenue in about a decade.

Revenue erosion big enough to remove half of revenue within a decade is roughly equivalent to a seven percent a year decline. So even if new sources grow three percent a year, losses still will happen.

To sustain revenues at their current level might therefore require annual growth of seven percent. That is not going to happen, in most markets. As James Sullivan, J.P. Morgan head of Asia equity research (all of Asia except Japan) telecom revenue growth is now less than GDP growth.

68 major telecoms groups – aggregate revenue, 2009-2016

source: Telco 2.0 Research  

Other analysts make the same argument, namely that revenue growth, at a global level, now is less than one percent.

Peak telecom is coming.

Why App-Access Provider Partnerships, Even When Necessary, are so Difficult

One often-cited bit of advice for mobile operators trying to change their value propositions is to wrap more functionality around their core offers. That does not mean "moving up the stack" (occupying new positions in the ecosystem) so much as enhancing existing offers. 

Sometimes mobile operators do want to move into different roles within the ecosystem, though. A common bit of advice is to partner to accomplish such objectives. Such deals can be low-margin gambits, though. 

For starters, profit margins typically are difficult, because there is a cost of goods issue, especially for services that already have low profit margins.  

Any business deal requires that all parties gain value from any transaction. So when any service provider wants to partner with another application provider, value has to exist for both parties.

You might argue that one reason tier-one service providers have not have much success partnering with voice or messaging application providers, to compete with independent providers of over-the-top voice and messaging apps, is that too little value generally exists for the potential app partner.

That is the logical consequence of the way we have decided to build the next-generation telecom network. Those of you with long memories will recall a number of such attempts.

One might argue early business data networks such as X.25 or frame relay were not intended to be universal solutions. But efforts to build general purpose next generation networks have been numerous.

Some will recall ISDN (integrated services digital network) and its successor, “broadband ISDN” (later known as ATM, or asynchronous transfer mode. Some of you also will recall the huge debates about whether sticking with ATM made sense, or whether the industry should shift to Internet Protocol.

Those debates were hot and heavy two decades ago, when many now in the industry were not born, or were youngsters. ATM lost and IP won, in large part for cost reasons. Ethernet connections were widely deemed to be ubiquitous and well down the cos curve, compared to ATM customer premises equipment.

There are some important consequences. Once IP becomes the next generation protocol, all applications essentially become “over the top” apps, no matter who owns the apps. That is worth emphasizing.

In an IP environment, no application provider “needs” a business relationship with an access provider. That, in turn, affects all potential partnerships between access providers and app providers.

Simply put, application providers need no business relationships with access providers to get access to their potential users and customers. Service providers, on the other hand, tend to be the entities that “need” partnerships.

Every successful business deal requires an exchange of value for all parties. In an IP environment, app providers do not “need” such relationships with access providers to reach their customers. That limits the value any access provider brings to any potential deal.

source: ZDnet