One thing seems clear, after a couple of announcements in recent weeks and several years in the doldrums: network slicing is coming, and it will make a new virtue of nationally-available 5G networks for enterprises. Last week, the analyst community debunked a few myths about slicing, even as the technology starts to gain commercial purchase – as sliceable standalone 5G (5G SA) technology replaces derivative non-standalone (5G NSA) systems in networks around the world. Is slicing ‘coming of age’? was the question; ‘you’re having a laugh, right?’ was the answer; in fact, the response rather challenged mobile operators to sell a solution, for once, rather than just a technology.
Because it does not matter how good a technology is (and slicing may be very good), it will never find its mark with enterprises if it does not solve their problems. That was the point the analysts made, and it is a perfectly familiar one, which is put to excitable vendors every time they pitch up with new toys and big talk about digital change. “No enterprise has a 5G problem; no enterprise has a slicing problem” – was probably the choice quote from the piece. It is the same with AI, IoT, and any incarnation of 5G: scope the problem, design the solution, and maybe there is a place for these things in the end. But mobile operators are just not great at this, the analysts implied.
At the same time, in response to questions from RCR Wireless, some of them sound measured, sensible, relaxed. Some have been testing the technology with customers through the first half of the year; some are readying it for commercial launch in the second half. But none of them sound like they are in a rush; none of them sound like they think they have it all worked out; none of them talk about slicing like it is more than just another tool in the 5G tool box. A spokesperson for Deutsche Telekom said: “We will continue to focus on innovation with our partners and customers to design and explore solutions leveraging 5G SA, including dynamic slicing technology.”
Yes, the focus is on technology; but, even here, it is solutions first. It may be something and nothing, but Deutsche Telekom also talks about the “potential value” of national network slicing for enterprises in the “areas of critical communications and Industry 4.0”. Again, the word choice seems significant; nothing is ruled in or out, as yet, but the sense is the industrial market may find a use for it through collaborative problem solving. As well, the Germany-based firm, notes the lesson about device-shortages from both ends of the IoT game – from the age-old low-power (LPWA) end, and from the trendy new high-power (private 5G) end.
“But we are dependent on the entire ecosystem to develop further, and [on] the availability of 5G SA devices,” it says. We are projecting, but it is like the carrier community does not want to be caught advertising an undercooked dish, again, even as the stove fires up and the seasoning goes on; slicing is still not on the main menu, yet. Verizon says the same, when asked for a timeline for a commercial proposition. “[It] is partly dependent on the greater ecosystem – as more smartphones and IoT devices get adopted into the marketplace, the commercial proposition advances. The 5G SA core, RAN, and device ecosystem has to evolve together,” says a spokesperson for the US firm.
“Our approach is to launch… only as those elements reach a viable state of maturity, so [they] deliver meaningful benefits.” But Verizon is “taking initial steps”, it says, pointing to a field demo for public safety, and another for live motorsports content. In the end these seem like clearer use cases (solutions) for slicing of national 5G infrastructure, like with trials for music festivals and boat races. Deutsche Telekom – which owns Hrvatski Telekom in Croatia, which has sliced its network for a greenfield port (one of the aforementioned announcements) – has already sliced its home network for broadcast services, used “regularly and very successfully” during the Euro 2024 football competition.
It is the “first operator in Germany with a commercial offer based on public 5G SA slicing”, it says, with some gusto. “We anticipate that demand… will evolve in the future to fulfil specific use cases with high reliability requirements or specific security demands for example,” it adds, with more reserve. Away from consumer-led broadband disciplines with content broadcasting/streaming, and also from more-abstract Industry 4.0 possibilities, it seems like national security and/or public safety is the tinder-box application for public 5G slicing. This is what Drei Austria is doing for blue-light services in its home market – which is the other new announcement referenced at the top of the piece.
“Media and entertainment and gaming are consumer-facing enterprise examples the industry is exploring, alongside public safety. We anticipate that public sector and public safety use cases will be among the earliest and most prevalent applications for network slicing,” says Verizon. As an aside, it is curious at the same time that Nokia’s mega drone fleet in Switzerland, deployed by Swisscom for national emergency and safety services, is using public 5G for its remote controls and data streams. Swisscom says slicing is on the cards, like private networks, possibly for other use cases, but that general-purpose 5G gets the job done, as it stands, for blue-light drone applications.
Asked in email about likely applications for public 5G slicing, France-based Orange writes: “When it comes to public networks, one target for 5G slicing is… to address critical use cases [such as emergency and security services] on a wide geographical area – an entire country, for example. Companies with… multiple sites, field technicians, or moving assets… are also potential candidates. For these companies, guaranteed bandwidth is key and a fully private 5G network on site is not the answer.” The point, then, is that high-reliability national coverage is the motivating factor, and applies where assets are scattered or in motion.
Orange makes the case that public 5G connectivity has relevance at the industrial edge, too – in fixed enterprise plants and campuses – in combination with a slice and a private/hybrid edge setup. But these Industry 4.0 locales will look to ring-fence operations with dedicated edge-5G first, it says. “We offer hybridization as an intermediate option, enabling access to wide public spectrum and bandwidth with related priority (slicing) [capabilities] while being able to keep data on site. Particularly large or sensitive industrial sites – ports, airports, industrial sites, chemical sites – are more likely to adopt 5G technology and 5G slicing on a hybrid or fully private network for their critical use cases.”
But private 5G is a point solution, essentially, even if it is strung-out across a number of locales as a multi-point solution; the public network, in some form, is required to chain-together supply-chain assets in-between. It is also relatively expensive, compared to publicly-available 5G, whether sliced, or annexed in hybrid-private mode using the public core network. Orange says: “Private 5G is very customizable, providing enhanced security, higher bandwidth, and lower latency … [But] it requires investments in base stations, antennas, and backhaul, for example…. [It also requires] dedicated spectrum and skilled personnel. [Whereas] public 5G is more accessible and cost-effective.”
Orange goes on: “[A hybrid of the two] offers the best of both… [It] provides the elasticity to run different operations on [different] networks [where] the private network takes care of business-critical functions… and the public network is used for less sensitive tasks. A hybrid approach [gives] the option to diversify networks to reduce the risk of disruptions and downtime. Enterprises can also reduce costs by utilising low-cost public networks for non-critical workflows – optimising network resources to meet their needs.” This is the takeaway from the discussion – that slicing is useful, potentially, in certain problems / solutions and important, probably as part of the whole mix.
Others say the same. “We are offering various products for enterprise customers to address their needs and help them to digitise their businesses. With our campus network proposition, we offer distinct products leveraging public and isolated private networks or navigating automatically between public and private. These options give customers choice to address their enterprise needs today,” writes Deutsche Telekom. Sure, it sounds a little scripted, but it also sounds open and realistic, perhaps. Verizon, meanwhile, gives a whole explanation of how all the hammers and tongs in its bulging 5G toolbelt may be variously used and combined – which is edited, copied, pasted below.
In response to a question about whether national 5G slicing is really for less-critical enterprise applications, thereby precluding its relevance for most of the Industry 4.0 market, where private 5G will probably prevail, Verizon writes: “[Public 5G slicing] has Industry 4.0 relevance and could be used for mission-critical comms, e.g. in public sector/public safety scenarios. Other commercial uses can be handled by private networks, neutral host networks, or through network slicing, depending on the needs of the client.
“For example, smart meters need little in the way of bandwidth and nothing in the way of mobility – but they do need a high degree of reliability. They require a more modest use of network resources [and could use a slice accordingly]. By contrast, massive multiplayer online mobile gaming requires high upload/download speeds and low latency… [So] that could be another slice. [Meanwhile] fixed wireless access (FWA) on a 5G network does not need mobility routing. Slicing could allow a provider to commit to minimum speeds for FWA. Each of these services would need a network slice with very different characteristics.”
Verizon goes on to explain slicing scenarios in a mixed-use campus, where the 5G supply might come via the public network, a neutral host network, or a private network. It says: “Office environments may be able to function with average throughput and latency, while video labs, factory floors, and other experimental areas might require higher performance. Slicing can allow each of these environments to have service performance appropriate to the needs of their applications. [It] allows the network to dynamically and efficiently define the experience a customer should get for the applications they are using, and provide that specific service level assuming resources are available.”
