Unlocking the problem of network slicing – what does this really mean?

Network slicing has, rightly, attracted considerable attention. It’s key to monetizing 5G investments. But, it’s not here yet – and that has caused problems. Why? When will we see slicing emerge from the shadows? And, what does this mean for service assurance?

Has the promise of slicing been exaggerated? It’s a matter of interpretation

It is a truth generally acknowledged that network slicing is key to meeting future revenue and growth from investments in 5G networks. While the term has been around for some years, it’s also true to say that network slicing has yet to gain the expected momentum – and, in some quarters this has led to skepticism regarding its potential role as an engine of monetization.

This is somewhat misleading – and misplaced – and may be due to a misunderstanding of what slicing actually means in the network. In this blog, we’ll explore the concept of slicing, with reference to our recent white paper “CX Management in 5G SA Networks”.

Understanding what 5G means – a multi-part story

Part of the problem arises because of how we interpret the term 5G. Sure, we have plenty of 5G networks now (more than 500 are commercially available, according to the GSA) – and counting. Data from Ericsson reports that these serve more than one billion subscriptions, so in terms of sheer numbers 5G looks like a success. 5G seems to be everywhere.

That’s only partly true. In reality, today’s 5G networks are largely (let’s be honest, overwhelmingly) based on the existing 4G core with the addition of 5G radios. This means that users have access to faster speeds and some latency enhancements – but are they getting true 5G?

The answer is no. That’s because, while 5G is designed to support multiple parallel services, this can only be realized with slicing. Today’s 5G networks do not (by and large – there are exceptions) support slicing yet, so the promise that this new technology offers has been pushed down the road a little. We’re still firmly anchored in ‘non-standalone’ 5G (5G NSA), which was brought to market fast for a variety of reasons – but this isn’t the 5G that was hyped so much back in 2019, for example.

The 5G that everyone was shouting about is really the second phase: standalone, or SA. It is only when we move to 5G SA networks that we will be able to unlock the slicing opportunity, that is, by enabling independent multiple networks to be created on a common physical infrastructure. 5G SA is designed to support multiple parallel services – Enhanced Mobile Broadband (eMBB), Ultra-reliable Low Latency Communications (URLLC), for example – but these are not deployed over a basic 5G NSA architecture like we see today; they are in fact slices that enable UEs to obtain certain performance requirements in the context of 5G SA.

In other words, what we have today with 5G NSA is something that delivers some of the performance requirements in the IMT definition of what constitutes a 5G service, but it doesn’t (cannot) deliver all of these, simply because it isn’t based on slicing.

It is only when 5G SA is brought to service that we shall see the introduction of slicing – and hence the availability of eMBB, URLLC and other such services. In our opinion, many have conflated NSA and SA, leading to a crucial – but subtle – conceptual misunderstanding that has led to confusion in some quarters regarding what 5G actually is.

Network slicing can only be realized with 5G SA

So, let’s recap. We have a version of 5G (NSA), but this does not support slicing. Soon, we will have widespread availability of 5G SA – sometimes called full-5G – which is inherently based on slicing. With 5G SA, all services “may be supported by one or a combination of network slices” as per 3GPP – in other words, slicing is yet to emerge, and we are some way from realizing the potential of 5G.

The hype may be real, but it was premature – and, because it created enormous expectations, it’s understandable that skepticism grew when that promise was not immediately delivered.

However, the promise is real, and we hope it will be unlocked – but only once 5G SA is generally available. The good news is that this is coming. Already, there are more than 36 CSPs that have launched or deployed public 5G SA networks and many more are expected in 2024 and 2025.

While this is a more optimistic perspective than you may have read from some media articles, it also brings our attention back to a pretty large elephant hiding in the corner somewhere: service assurance in a world of slices.

Slice performance targets and parallel slices introduce new levels of complexity

That’s because each 5G slice (the original set of three, plus others that have since been defined by 3GPP) has different performance targets, as we noted in our recent paper. To give just one example, URLLC is usually characterized by two key demands, as originally defined in the IMT:

• Network reliability >99.999% and even up to 99.9999
• Latency of around 1ms for data transfer

In terms of service assurance (and customer experience – the focus of our paper), we have the complication that we must consider a specific service (e.g., multi-role online games) delivered across a given slice (e.g., eMBB), alongside other slices and services that are also active at the same time.

eMBB, URLLC and the rest will coexist in the network. In turn, these MNO slices may also coexist with slices dedicated to a particular user group – an eMBB slice offer exclusively reserved for a particular MVNO, for example, or a URLLC slice for a smart port, autonomous factory, or something else – supporting the expected B2B cases.

Read our paper to explore service assurance in a sliced world

The future will herald multiple, parallel slices. Some slices will be for large numbers of users and devices, while others may have only a few – a trend already encountered by Elisa Polystar. All of this must be catered for by service assurance platforms.

To understand what that means and how we can tackle this problem – which is central to the task of monetizing 5G successfully – read our new white paper, in which we explore service assurance and customer experience in the context of a slice-enabled 5G world.