We’ve heard about 5G network slicing for a while now and although testing has progressed, the industry is yet to make large deployments.  But why, and when can we expect the technology to hit the US? DIANA GOOVAERTS digs into this.

There are two key things to know about network slicing. First, it isn’t necessarily a new concept and, second the type of implementation most are talking about in a 5G context requires a standalone (SA) core.

Verizon SVP of technology planning and development, Adam Koeppe told Mobile World Live (MWL) partitioning the network to route traffic a certain way based on device requirements has been around since the early days of LTE.

He said the difference with 5G is “you’ve now got a fully programmable virtual infrastructure that has automation and orchestration on top of it, and you can create more slices in a much more elegant way than you could in the past”.

This “is really designed for the 5G SA core. Those two go hand-in-hand”.

Verizon is working toward “full commercialisation” of its SA 5G core in H2 2021 and, once complete, Koeppe said the operator will be able to deploy slicing if and where it needs.

AT&T told MWL it began initial SA deployments in late 2020 and is “on track to scale and deliver capabilities to our customers when they need it”.

T-Mobile US launched nationwide SA 5G coverage in August 2020: SVP of radio network engineering and development, Mark McDiarmid told MWL the operator is working hard with vendor partners to develop slicing capabilities, and mentioned home broadband as a potential contender for early application of the technology.

Broadband is a bit easier because “the devices and homes are not mobile so you don’t need to include the mobility functions in 5G, in the core network when you’re serving that type of session”.

He added the operator could realistically deploy the capability “in the next 12 months”, though a T-Mobile representative later noted this was not a commitment to do so.

Drivers

Koeppe stated network slicing deployments will largely be driven “by an operator’s need to serve traffic efficiently”.

“It’s not like a consumer use case or enterprise use case is saying this can only function with network slicing”.

But, in a report issued last month, ABI Research highlighted 5G slicing as a key enabler of new revenue streams for operators, estimating the technology would generate cumulative revenue of more than US$20 billion globally by 2026. The majority of this (US$12 billion) was tipped to come from serving industrial manufacturing, C-V2X and logistics segments.

Meanwhile, GSMA Intelligence’s Operator in Focus 2020 survey found operators view automotive manufacturing, financial services and healthcare as the top three industry verticals which will benefit the most from network slicing.

It noted 22 per cent of operators are likely to pitch network slicing as the primary 5G value proposition for enterprises over the next two years.

ABI Research outlined three factors driving operator interest in network slicing: the ability to optimise network efficiency; to rollout new services without disrupting existing ones; and to help enterprise customers offer a wider range of services.

Senior analyst for 5G core and edge networks, Don Alusha stated “the industry at large now realises that to extract the value at stake, there is a need to enhance the traditional way of doing business and clearly articulate business drivers and commercial utility of slicing to vertical partners”.

Other factors

Interest in slicing is one thing, execution another.

Beyond the need for a SA 5G core, there are a several other considerations operators must consider, including device compatibility and roaming.

Koeppe explained a SA 5G core will also support non-standalone (NSA) applications, but noted a large proportion of early 5G devices are only compatible with NSA. As a result, Verizon (and likely others) will need to consider whether there is “a use case on NSA devices that makes sense to fit into the slicing capabilities”.

“We may have one and we’ll implement that, but at this time it’s to be determined.”

Roaming and the ability to implement slicing across operators is another factor. Koeppe noted 3GPP’s 5G specifications will define “the basics of what a network slice is”, making it possible for two different operators with SA cores to create matching slices.

Issues may arise, however, if a customer roams on a network less advanced than their home network.

“It creates a lot more opportunities for different types of roaming relationships, but it also adds an element of complexity,” he said, adding it is still “very early for that type of implementation”.

Alusha concluded the path toward network slicing is a “long-haul endeavour” which will likely begin with small, single-vendor implementations.

Multi-vendor deployments “will materialise with further industry alignments on terminal support, business model, and collaboration among system integrators, vendors and CSPs,” he said.

• Editor’s Note: The editorial views expressed in this article are solely those of the author and will not necessarily reflect the views of the GSMA, its Members or Associate Members.