Is the sweet spot for 5G SA disappearing?

David Martin, senior analyst and head of telecom cloud at STL Partners, told Fierce that while "a lot of promises" have been made by operators for 5G SA deployments around 2021 and 2022, many of those promises have yet to materialize.

"Operators have been almost completely silent on this," Martin said. We came to the conclusion that, in reality, many [of the planned deployments] will never be completed." According to STL Partners, this is due to a number of different factors.

As Martin explained, operators may have been delaying the deployment of 5G SA due to the uncertainty surrounding the SA deployment itself, coupled with a lack of confidence in deploying 5G SA on the public cloud. "It's kind of a vicious circle, in the sense that SA is a network function that is well suited to be deployed on the public cloud, but operators are understandably very uncertain about the broader implications of doing so in terms of regulations, performance, security, resilience and so on," Martin said. Martin noted that greater confidence in 5G SA use cases could drive more operators to deploy them on the public cloud. However, he said, beyond the potential of network slicing, "very few useful cases have been developed and commercialized."

In addition, operators are already struggling to generate returns from existing investments in non-standalone 5G (5G NSA). STL also highlights changes in the public cloud providers themselves. It noted, for example, that there were doubts about Microsoft's commitment to the telecom cloud after it restructured its carrier business to include mobile core products including pre-discontinued Affirmed and Metaswitch product sets. "I think this is causing operators more hesitation because AWS is well positioned to take advantage of this opportunity and establish leadership and dominance in public cloud-enabled network capabilities, but operators clearly don't want AWS to dominate and they may have to wait until other players scale up and demonstrate the performance and resiliency of their cloud infrastructure," Martin said. He pointed to Google Cloud and Oracle as two vendors that could "fill the gap." Another reason for the hesitancy about 5G SA is that some operators may now be looking for newer technologies such as 5G Advanced and 6G. Martin said the 5G Advanced(also known as 5.5G) use case does not typically need to be used in isolation, but he noted that RedCap technology is an exception because it relies on 5G SA's network slicing and large-scale machine-type communication (or eMTC) capabilities. "So if RedCap is adopted more widely, it could act as a catalyst," he said.

Editor's note: Following the publication of this article, Sue Rudd, managing director of BBand Communications, said 5G Advanced has always required 5G SA as a prerequisite, not just RedCap 'with an exception'. "All standard 3GPP 5G advanced features leverage a 5G service-based architecture," she said. At the same time, Martin observes, many operators are now at the end of the 5G investment cycle, and "they're going to start looking at 6G." Martin noted that Tier 1 operators that have already rolled out 5G SA at scale "will now seek a return on these investments by developing network slicing use cases," but he said that "a long list of operators that have yet to launch 5G SA may now wait on the sidelines, perhaps simply exploring 5.5G and delaying SA deployments indefinitely."

At the same time, the STL report suggests that the prospects for vRAN and open RAN look more promising than 5G SA, where vRAN is defined as compliant with Open RAN standards but is typically offered by a single vendor. Here, Martin makes it clear that operators do not have to synchronize investments in 5G SA and vRAN/Open RAN, and that one investment does not necessarily predetermine the other. At the same time, he said operators have been unsure which of the two investments should be prioritized, and they are questioning whether 5G SA is really needed to "fully leverage the benefits of Open RAN, especially in terms of RAN programmability for network slicing and spectrum management." This is also a complicating factor. "I think operators have been thinking about these questions for the last two or three years, not just about SA, but how do we treat the public cloud? Are we going to adopt a fully multi-cloud model?

All of these issues are interconnected, and you can't look at any one of them in isolation and ignore the big picture, "he added. STL's report notes that in 2024, significant Open/vRAN projects from major operators including AT&T, Deutsche Telekom, Orange and STC are expected to begin commercial operations to some extent. Martin added that the vRAN model "has the potential to be a successful model for 5G open RAN." "Many factors still need to come together, including performance, cost, energy efficiency and the ability to demonstrate its deployment in an open manner." But I think the potential of vRAN is very big, "he said.