Ask the Expert: What Is 5G Network Slicing? And How Can It Improve My Business Applications?

Though we hear about the speed and latency benefits of 5G, there are lesser-known features such as “network slicing” that warrant more consideration by operational, engineering and IT decision-makers.

A warehouse worker uses a Zebra ET8x rugged tablet when inspecting inventory
by Your Edge Blog Team
April 08, 2022

All you have to do is turn on a TV to see commercials touting the “speeds & feeds” benefits of 5G. Even your next-door neighbor is probably talking about how this next-gen cellular technology is going to change the world, or at least their media streaming potential. But as enterprise mobility expert and Zebra Engineering Fellow Bruce Willins has reminded us many times before, the business benefits of 5G are quite broad and rapidly evolving to meet enterprise demands.

There are many functional benefits emerging from the 5G transition that most organizations have yet to recognize, he said in a recent touch base. So, we wanted to share those with you ASAP because, let’s face it, everyone wants a greater return on investment (ROI) from technology and there’s also a compelling interest in technology-powered strategies and solutions that deliver a performance edge. 

Here’s what you need to know about 5G network slicing:

Your Edge Blog Team: Can you explain in the simplest terms what 5G network slicing is and why everyone from IT leaders to operations managers should be looking into it right now?

Bruce: Let’s start with an analogy to parcel delivery where you have various service level options such as expedited delivery, high reliability, high security, temperature controlled and the like. Now imagine you have a similar situation in which 5G traffic can be sent with different service levels, each associated with a “slice.” Each slice is a virtual private network, customizable to specific enterprise needs and capable of covering a large geographic area. So, slice #1 might be for augmented reality (AR)/virtual reality (VR) or video streaming, slice #2 might be for low-latency applications such as IOT sensors, slice #3 for high reliability, and so on.  Slice assignments and slice attributes can be dynamically adapted to meet changing business needs and to more efficiently utilize network infrastructure resources. 
 
If you think about it, cloud is to computing as network slicing is to networking.  Whereas cloud enables efficient, scalable use of compute resources, slicing does exactly the same for networking. Slicing is fundamental to enabling Networks as a Service (NaaS), which is a key technology for hyperscalers like Google, Amazon, and Microsoft. The NaaS market is forecast to be $45 billion in 2026, growing at a CAGR of 33.05% from 2021 to 2026. For those interested, Cisco has a good blog post on NaaS basics.

Your Edge Blog Team: So, is network slicing one of many features emerging from the transition from 4G to 5G?

Bruce: Absolutely. It is an important feature that comes from 5G devices operating over a 5G standalone (SA) infrastructure. It has been described as “the breakthrough technology for 5G.” Organizations can now have a customized, scalable network solution without having a capital investment outlay and/or having to manage a cellular network infrastructure.  

Your Edge Blog Team: Do Zebra devices support network slicing?

Bruce: Yes, all Zebra 5G devices are capable of supporting network slicing. Though it’s important to note that network slicing is optimally implemented over 5G SA network infrastructure. While nearly all major carriers worldwide have initiated 5G SA rollouts, it is not yet ubiquitous.

Your Edge Blog Team: Does Android™ include native support for network slicing?

Bruce: Yes, Google added rudimentary network slicing in Android 12. I can’t say anything for certain yet, but I expect that additional network slicing support will be included in Android 13, which will go open source in the second half of this year.

Your Edge Blog Team: We’ve talked about private networks in prior discussions. Is 5G slicing a private network?  

Bruce: These are both complementary and overlapping technologies. You can think of a network slice as a “virtual” private network. Just to recap a bit, today most enterprises operate on a shared public network. Data privacy is marginal, and throughput/latency performance is unpredictable – you get what you get. This has driven some enterprise customers to deploy private wireless networks. Private wireless networks are not new to 5G, they have existed on 4G for quite some time. Industry analysts estimate that the market for private LTE networks in 2018 exceeded $1 billion. 
 
As the name implies, a private network allows an enterprise to own and/or operate their own “private” cellular network. They can do so organically or by leveraging a service provider. Taking ownership of the physical hardware (customer premise equipment, or CPE) does require a capital outlay. Operating the network implies that the enterprise has technical know-how and resources to do so. For customers retaining their Wi-Fi network, this may mean managing two disparate networks and addressing roaming requirements. Finally, these networks can be operated in licensed or unlicensed spectrum. If using licensed spectrum, the enterprise must either obtain spectrum via auction or lease spectrum from a third party. If operating unlicensed, then an analysis must be made of available spectrum in the area of deployment. As we’ve discussed before, such solutions may not scale well over a large geographic area and are therefore commonly relegated to confined areas, such as a campus, stadium, or airport.  

Thus, 5G slicing offers many of the same features customers seek in private networks – like data privacy and scalability but can operate over wide areas, offer added granularity for application and traffic profiles, and offer end-to-end service levels (i.e., inclusive of the Radio Access Network, or RAN). 
 
By no means are private networks and network slicing mutually exclusive. In fact, many believe that a hybrid model connecting localized private networks over a sliced public network will be a popular configuration.

Your Edge Blog Team:  It seems like part of the value of slicing is the mapping of traffic onto slices with different service characteristics.  How does 5G address this mapping?

Bruce: It’s a bit complicated, but its standardized and flexible. The mapping process is coordinated between the infrastructure and the user equipment (UE). Traffic/applications are assigned to a specific slice based on a dynamically configurable set of User Equipment Route Selection Policies (URSP) provisioned by the infrastructure Point Control Function (PCF). Using these policies, the UE routes traffic to the prescribed slices.

Your Edge Blog Team: Is there anything in 4G comparable to 5G network slicing?

Bruce: It is possible to get a subset of network slicing functionality using a 4G private access point name (APN).  In general, an APN is used to setup a UE mobile network connection to a packet gateway (P-GW). The packet gateway (P-GW) then connects to a packet data network (PDN) such as the public internet. In a typical consumer model, the user is now connected to the internet.

In an enterprise environment, a private APN is typically used to connect remote users to a corporate network.  Configured with a private APN, authorized UE devices connect to an ingress gateway (P-GW) which establishes an IP SEC tunnel connection (i.e., across the public internet) to an egress P-GW which connects to the corporate network – thus, securely connecting corporate users to corporate resources. Though the private APN segregates and securely tunnels traffic across the public internet, it has no quality of service (QoS) guarantees. 

Your Edge Blog Team: Can you provide a bit more color on how slices might enhance security?

Bruce: Security is often about isolation and containment. It is easier to contain a cyberattack within a single network slice, so security is certainly a benefit. Slices also provide logical data privacy and isolation from public traffic. 

Your Edge Blog Team: By eliminating capital outlays for infrastructure, does network slicing – and NaaS – make 5G more accessible to enterprises? Or perhaps more scalable?

Bruce: Dynamic scaling is definitely a functional benefit of 5G network slicing, as is the ability to customize service levels without the need to license or lease spectrum or operate in a limited-capacity unlicensed band. Organizations can get broad 5G network coverage yet tailor the service experience to user groups or applications. It’s more about increased customization rather than increased accessibility, because the 5G services are being delivered via the same public wireless infrastructure. The network is just being sliced to allow for the delivery of tailored services to different user groups within the same customer account (or organization). It also allows for greater overall efficiency of network infrastructure resources.

Your Edge Blog Team: Who “owns” network slicing? Is this something handled by IT departments? Does the burden fall to communication service providers (CSP)?

Bruce: It is expected that carriers or CSPs will sell capacity in the form of slices based on the specific needs of enterprise customers, which would essentially be a virtual private network. Customers will be charged based on their specific service level requirements.

Your Edge Blog Team: Are network slicing capabilities motivation enough for many companies to speed up their migration from 4G to 5G?

Bruce: It’s more about preparedness than acceleration. Our studies have shown that over 58% of enterprise customers expect to keep their devices in service for five or more years. That means the majority of enterprise devices purchased today will be in service on or after 2027. Projections are that, by 2025, network slicing will be a $1.3 billion market with a heavy emphasis on enterprise verticals. A report published by Ericsson shows that the top five verticals driving network slicing revenues for CSPS will be Healthcare (21%), Government (17%), Transportation (15%), Energy & Utilities (14%), and Manufacturing (12%).  This same report states, “25-30% of the potential 5G use cases will need slicing as an enabler.” 
 
From a big-picture point of view, you can now see how cloud computing combined with NaaS (enabled by slicing) forms the basis for Infrastructure-as-a-Service (IaaS) which is the focus of hyperscaler managed services.  

Your Edge Blog Team: So, if you had to sum it up, what are the advantages of 5G network slicing?

Bruce: Customers making a multi-year investment in enterprise-class devices should consider both the tactical and strategic benefits of 5G. Speeds and feeds are easy to understand, but strategic aspects like network slicing are complex and less publicized.  

Enterprise customer use cases and security requirements are outgrowing the current public shared network model. Private networks are filling some of the gap, but still have issues such as wide area coverage. Network slicing fills many of the gaps. Although slicing is still in its early stages, widespread adoption falls well within the service life of currently procured devices. 
 
In summary, the future of network slicing is quite positive. Whether your enterprise is leveraging a centralized cloud computing model or an edge computing model, the effectiveness of these models is largely dependent on having the right level of end-to-end network service. This plays well into the 5G network slicing value propositions:

1).  Ability to create slices with customized service levels

2).  Eliminates customer capital outlays for infrastructure

3).  Eliminates customer IT administrative overhead 

4).  Provides logical data privacy/isolation from public traffic

5).  Cyber-attack containment within a single slice

5).  End-to-End Service Provisioning from UE to UE (inclusive of the Radio Access Network-RAN)

6).  Dynamic scaling to meet changing requirements

7).  Broad geographic coverage

9).  No need to license or lease spectrum or to operate in a limited capacity unlicensed band

10).  Ability for CSPs to offer network services tailored to enterprise vertical use cases

11).  Overall efficiency of network infrastructure resources

Make no mistake, Zebra has made a significant investment in 5G technology. We did so knowing full well the tactical and strategic benefits of 5G. Our goal now is to make sure that our customers have that same degree of awareness so they can make the best decisions for their businesses.   

###

Related Resources:

null
Your Edge Blog Team
We want to hear from you! Submit your comments, questions and topic ideas to blog@zebra.com.