What is the Business Value of a Network Slice?

What is the business value of a network slice, the ability to create an end-to-end virtual network with features optimized to fit particular use cases on a number of dimensions?  Network slicing is the logical partitioning of a physical network into independent virtual mobile networks.

So the issue is what value various mobile operators will see to creating a multi-tenant capability. Should that extend only to the operator’s enterprise customers, or should multi-tenancy also extend to enabling rivals (as traditionally is the case for network operators selling capacity to third party competitors who want to operate mobile virtual network businesses?

The easiest case, for a network operator, is using network slicing to better support its own retail customers.

For starters, network slicing implies the ability to tune the core network features to match different use cases.

An automated vehicle control network might emphasize ultra-low latency. An industrial internet of things network might emphasize low-bandwidth and infrequent communications.

A high-end consumer internet access use case might focus most on bandwidth. Health applications might value reliability and predictability above all else.

Existing consumer mobile and fixed network services are “one-size-fits-all.” That has operating and capital cost implications since most use cases arguably result in network resources being underused (stranded).

On the other hand, some use cases might require either high-bandwidth or low-latency, but not both. Some use cases benefit from higher levels of security; others are less stringent.

So network slicing offers the hope of more efficient networks that also can be customized, to an extent, to fit specific use cases.

Each network slice can be optimized to provide the required resources and class or service or quality of service to meet the diverse set of requirements for different use cases.

Source: CableLabs

There could be other far-reaching implications, however. What network slicing could mean is “multi-tenant” use of a fixed network access facility. In other words, it should be possible for multiple independent users (enterprises, app provider or service provider) to manage their own networks over the one physical access link.

Think of the ways mobile virtual network operators now provision network resources. To be sure, capacity will be sold wholesale in some way that relates to network usage (bandwidth, for the most part). Today, MVNOs pay their suppliers based on consumption (volume).  

With network slicing, other possibilities can be envisioned. Perhaps an MVNO buys a fixed slice of capacity and use (fixed or flat fee) and then is able to tweak functionality to match its expected use cases. This might be particularly useful for a specialized MVNO serving customers with special requirements (health apps and services emphasizing class of service protections).

Perhaps an autonomous vehicle network really wants ultra-low latency, above all, and builds that into its own slice.

Also, at least in principle, some entities could consider co-investment in fiber to home facilities that share one physical link but are partitioned into multiple virtual connections. That could be a substitute for today’s practice of buying standard capacity, with charging based on usage, and a new way to share infrastructure costs without sacrificing “control and differentiation” of network features.

Put another way, could network slicing be a new way of allowing multiple service providers to share the cost of access infrastructure? Traditionally, service providers have shared physical elements such as towers, radios, metro transport or access facilities. Sharing of towers and radios has been somewhat common, in some markets, and will be a feature of South Korea mobile 5G infrastructure, for example.

Nations with a wholesale approach to facilities have had one physical network supplier and then multiple retailers using that common infrastructure.

But network slicing in the 5G era could bring substantial changes. Perhaps network slicing creates a new way for mobile operators and others to share the costs of building networks, perhaps most obviously for the half of radio sites that typically carry only about 10 percent of traffic.

Since executives are likely to want to maintain full control over any elements of their business that provide customer-facing strategic advantages, one possible driver of network slicing is to use that technique to reduce the cost of coverage for up to half of sites that carry very little traffic, and then maintaining full control over the half of radio assets that represent those portions of the network carrying 90 percent of traffic.

As always, such practices might appeal most to non-dominant or smaller service providers, since the capex savings could be significant over the approach of building an entirely-owned network. Network slicing would save the costs of negotiating site leases, not simply the cost of facilities, for example.

While useful for traditional MVNOs, the advantages are even clearer for specialized vertical-oriented MVNOs operated by non-traditional providers, who might see network slicing as a way to source a full network, instead of acquiring capacity the traditional way.

Major app or service providers whose main business is not “general purpose communications might well see many advantages to doing so.

Different actors will have differing views of the value of such arrangements. Dominant providers with leading market share might continue to favor retail operations of their own, and favor network slices for enterprise customers running their own private networks, over enabling competitors in the retail general mobile service markets.

But in some markets, new entrants might envision market entry as resource brokers, with what is essentially a “wholesale-only” model, something quite rare in the mobile business so far. In the U.S. market, that might fit some suppliers such as the proposed network being built by Dish Network.

Were that network to become a sustainable operation (if the assets are not sold), it is conceivable the network could adopt a “wholesale only” platform, using network slicing or traditional methods to support retail customers.

How Many Consumers Would Willingly Buy Internet Access at $80 a Month?

With news that the U.S. Federal Communications Commission is once again going make its annual study of U.S internet access speeds, keeping the minimum at 25 Mbps, it is worth noting that, in the first two quarters of 2018, the “mean” (average) mobile internet access speed was 27.3 Mbps, according to Ookla.

In the first two quarters of 2017, the mean fixed network speed was 64 Mbps, according to Ookla. With the normal caveats (mean is an “average”; median is the midpoint).

The point is that the FCC definition is a minimum definition, not a statement of direction, as both mobile and fixed network speeds, “on average,” already surpass that level.

Keep in mind that internet service providers operate on many different platforms, and not all the platforms can upgrade speed as “easily” as fixed network operators. One thinks of satellite operators, at least one of which would, were the minimum raised to 100 Mbps, as some already argue is needed, would not longer be able to sell a “broadband” service.

In principle, the FCC is trying to maintain a platform-agnostic metric.

To be sure, there are areas with what most would consider unacceptably slow fixed network speeds (below 10 Mbps), especially common in rural areas.  

And many would argue prices are too high. The point is that U.S. ISPs are making rapid progress on speeds available to most potential buyers.

The FCC report does not stake out new ground on aspirational levels of speed; that is happening in the market, in any case. What the FCC wants to do is set minimum expectations that all the platforms can reasonably expect to deliver, not the “maximums.”

source: PCmag

source: Ookla

By way of comparison, advocates of adopting a minimum definition of “broadband” at 100 Mbps might not be so clear on consumer willingness to pay for such speeds. One recent study suggests demand for 100 Mbps, at current pricing levels of roughly $65 a month, is close to the maximum most consumers are willing to pay.

As much as 75 percent of all consumers might refuse to pay that much for an internet access service.

Consumers seem to evaluate internet access offers the same as any other product, which is to say they compare value to price. And that means most consumers do not buy the most-expensive, fastest service, but tend to pick services that offer high enough value, at a reasonable enough price. In other words, they tend to buy the good enough service because it costs less than the “best possible” service, sold at premium prices.

“We find that households’ valuation of bandwidth is highly concave, with relatively little added value beyond 100 Mbps,” researchers assert in one study. “For example, households are willing to pay about $2.34 per Mbps ($14 total) monthly to increase bandwidth from 4 Mbps to 10 Mbps, $1.57 per Mbps ($24) to increase from 10 to 25 Mbps, and only $0.02 per Mbps ($19) for an increase from 100 Mbps to 1000 Mbps,” the authors say.

“As with the PC, one could again reasonably ask whether bandwidth increases continue to generate substantial value for consumers,” researchers at the Technology Policy Institute say. “Consumers highly value bandwidth enhancements at lower speeds, but the incremental value of bandwidth decreases rapidly” as speeds increase.

Since few consumers think about their internet access purchases in “dollars per Mbps,” the more relevant graph is that which shows willingness to pay for bandwidth at various speeds. Though total demand continues to increase from about 150 Mbps up to a gigabit, most of the demand is at 150 Mbps and below, TPI surveys have found.

source: Technology Policy Institute

One way of parsing this data is that about half the total demand is for service costing about $40 a month, while the overwhelming majority of consumers seem resistant to pay more than about $70 a month.

As you might expect, consumers value bigger usage allowances and lower latency, and appear willing to pay some incremental fees to get reduced latency and bigger usage allowances. Still, there also is reluctance to spend more than $80 per month for service, no matter how fast the service is.

Consumer Willingness to Pay for Gigabit Has Sharp Limits

Consumers seem to evaluate internet access offers the same as any other product, which is to say they compare value to price. And that means most consumers do not buy the most-expensive, fastest service, but tend to pick services that offer high enough value, at a reasonable enough price. In other words, they tend to buy the good enough service because it costs less than the “best possible” service, sold at premium prices.

“We find that households’ valuation of bandwidth is highly concave, with relatively little added value beyond 100 Mbps,” researchers assert in one study. “For example, households are willing to pay about $2.34 per Mbps ($14 total) monthly to increase bandwidth from 4 Mbps to 10 Mbps, $1.57 per Mbps ($24) to increase from 10 to 25 Mbps, and only $0.02 per Mbps ($19) for an increase from 100 Mbps to 1000 Mbps,” the authors say.

“As with the PC, one could again reasonably ask whether bandwidth increases continue to generate substantial value for consumers,” researchers at the Technology Policy Institute say. “Consumers highly value bandwidth enhancements at lower speeds, but the incremental value of bandwidth decreases rapidly” as speeds increase.

Since few consumers think about their internet access purchases in “dollars per Mbps,” the more relevant graph is that which shows willingness to pay for bandwidth at various speeds. Though total demand continues to increase from about 150 Mbps up to a gigabit, most of the demand is at 150 Mbps and below, TPI surveys have found.

source: Technology Policy Institute

One way of parsing this data is that about half the total demand is for service costing about $40 a month, while the overwhelming majority of consumers seem resistant to pay more than about $70 a month.

As you might expect, consumers value bigger usage allowances and lower latency, and appear willing to pay some incremental fees to get reduced latency and bigger usage allowances. Still, there also is reluctance to spend more than $80 per month for service, no matter how fast the service is.

Will Network Slicing Become a Substitute for Access Networks?

We often think of network slicing as a capability to be used by 5G networks to create virtual private networks all the way through the core of the network and sustained at the radio edges as well, so that VPN features flow through to end user devices.

In principle, though, network slicing will have to flow through fixed network edge devices as well (fiber to home and cable modem, for example).

Vodafone and Huawei have demonstrated network slicing over a fiber to home access network, for example.  

What network slicing could mean is “multi-tenant” use of a fixed network access facility. In other words, it should be possible for multiple independent users (enterprises) to manage their own networks over the one physical access link.

Also, at least in principle, some entities could consider co-investment in fiber to home facilities that share one physical link but are partitioned into multiple virtual connections. That would be an extension of the present practice in some countries of having one network services wholesaler and then any number of retail service providers operating over the single infrastructure.

Whether service providers in any particular market would consider such a move hinges on their perceptions of strategic advantage, network ownership and control, and the business model advantages.  

It might seem far fetched to expect U.S. cable operators, AT&T or Verizon to ever deliver their mission-critical retail apps over leased local access infrastructure. Smaller providers might reach different conclusions.

On the other hand, when network slicing gets to be nearly ubiquitous, the decision matrix could change, at the margin. Larger tier-one service providers operating their own big backbone networks, virtualized and with network slicing, might have little need to rely on third-party access facilities in region, or across their own mobile infrastructures.

Some argument might be made that a network slice makes sense out of the fixed network region, if and when the mobile network cannot itself do this. But those cases should ultimately be quite rare for the likes of Verizon and AT&T. It is not yet so clear whether such options will make more sense for Sprint, T-Mobile US or other mobile providers.

Fixed network providers operating out of region will have to look at the business case for buying a slice, versus building or leasing access.

The point is that access network strategy will have more variables as network slicing becomes a  commercial reality.

TDS Gets 46% Internet Access Market Share as an Overbuilder, in One Year

The key to success for any overbuilder attacking an internet service provider territory served by an incumbent telco or cable company is to pick markets where those providers are behind the curve in terms of speed. That is true for independent providers of all stripes and also seems to be the case for TDS, which now is overbuilding outside its core footprint.

In Sun Prairie, Wisconsin, where TDS has been operating for about a year, TDS has been able to achieve 46 percent broadband penetration and 24 percent take rates for video. In fact, says Vicki Villacrez, CFO for TDS Telecommunications Corporation, “our take rates are exceeding our experience in some of our more attractive ILEC fiber markets, where we are the incumbent telco company.” That is a big deal.

What those figures suggest is that TDS has been able to grab nearly half the market for internet service in about a year, a huge share of market  for an overbuilder.

As a result, TDS is looking to build fiber access networks in parts of Madison, Wisc.  and five communities near Madison.

“On a combined basis, these markets represent roughly 18,000 service addresses, targeting both residential and commercial services,” says Villacrez. Pre-sales activity in those new markets “has reached its pre-registration targets in three neighborhoods already and has reached over 50 percent of our required threshold to ensure we earn a targeted return on investment.”

China Mobile IoT Connections Grow 155 Million in First Half of 2018

With the caveat that one has to be careful about extrapolating too much from a single year’s financial results, or even a half year’s worth, China Mobile financial results so far in 2018 provide an excellent illustration of why mobile service providers are pinning so much hope on internet of things and other “non-human user” revenue sources.

In the first half of 2018, China Mobile has seen communications service revenue dip 5.5 percent, year over year, with mobile average revenue per user dipping 6.6 percent, year over year.

“In the first half of 2018, our IoT business achieved rapid growth, with a net addition of 155 million IoT smart connections,” China Mobile says. “The total number of IoT smart connections has reached 384 million and revenue from IoT business recorded a year-on-year growth of 47.6 percent.”

Network Slicing Returns to Some Elements of Debate Over "Stupid or Smart" Networks

Network slicing, an attribute of virtualized core networks, is in many ways the latest twist in thinking about where communications network intelligence should reside. Significantly, network slicing also is foundational for thinking about what can be done with 5G networks, and why 5G platforms will be different from all earlier mobile generations.

What one can do in the core of the network fundamentally drives what services can be created and offered at the edge of the network, of course. But in the IP era, that has been complemented by a rival approach, namely creating services and value at the edge of the network.

The best examples are any “internet” apps that are created by a combination of end user demand at the edge (from personal devices) and supply by data centers likewise at another edge of the network. In other words, cloud computing is the best example of value, apps and services created entirely at the edges of any communications network, and not in the core of the network, by the network itself.

But some changes are coming, at least for communications network operators. As core networks are virtualized, and as 5G gets deployed, it will be possible to create, as a principal network function, virtual networks that can be optimized on a number of dimensions to provide differentiated features.

Quality of service, latency performance, geographic reach, security, degrees of mobility and other features of the network service can be optimized for particular use cases. The shift is from one network with a standard set of capabilities to essentially multiple networks optimized, to an extent, for the use cases.

One use case might lean heavily on latency performance; another might require predictable packet arrival; throughput or security. The promise of network slicing is that, in principle, custom networks can be created.

Some of you, with long memories, can remember heated debates in the late 1990s and early 2000s about the “right” architecture for communications networks.

source: GSMA

Telecom professionals argued for “smart networks” with intelligence in the core, while those in the data communications camp argued for “dumb networks” with intelligence at the edge.

Some of you will remember this as the debate over whether asynchronous transfer mode (ATM) or IP should be the protocol used by the next generation network. Perhaps the best example is the argument made by David Isen that the future network should be a stupid network, putting intelligence at the edge, as IP devices do.

That sort of thinking got a heated response. But, in a fateful decision, everybody decided IP was the better choice, as a practical matter. No matter the invective, the “stupid network” became the preferred network for data, even on “telco” networks.

But the basic debate never dies down completely. New thinking could arise, at least for telecom architects, as networks are virtualized. In principle, that is a move in the direction back towards core network intelligence.

One might argue that all that is involved with NFV or software defined network approaches is where intelligence is situated (in fewer network elements) or who supplies those elements (specialized or general purpose; branded gear or white box platforms).

But there always are other dimensions. Firms that specialize in supplying IP bandwidth often fall in the “dumb network” camp, by definition: their product is, in fact, simple IP bandwidth.

"I'm much more in the 'not only dumb pipe, but dumb network' camp," said Dave Schaeffer, CEO of Cogent Communications Holdings.

Virtualized networks needed to support 5G, on the other hand, supporting network slicing, for example, shift back towards smart networks, by definition, as the network itself has to act to create virtual private networks as a core function. Up to this point, VPNs have functioned as “higher in the stack” overlays on IP transport.

Network slicing makes creation of virtual networks a core network capability. It is not exactly a return to the completely “smart” or “intelligence in the core” philosophy, but it is a significant step in that direction, without the overhead and complexity of legacy telecom networks, one might hope and expect.