Accelerating the network: Lessons from Lumen and SK Telecom

STL Partners has conducted extensive research on telcos globally, focusing on their edge computing deployments. STL’s report “Telco network edge computing: Lessons from early movers” delves into the experiences of Lumen, SK Telecom, Telefónica, Verizon, and Vodafone, all of which commercialised edge nodes before 2020. This article highlights key findings and four crucial insights gleaned from these early telco-edge movers.

Through our analysis, Lumen stands out as the most successful of the early edge movers. Slightly different to the others as an access-independent service provider, its success can be attributed to strong investment in its edge infrastructure, deployment of edge-enabled enterprise and internal services, and commitment to solving customer problems.

Network edge computing market is still nascent

Despite its potential, the network edge computing market demand growth has been slower than expected and the market was still nascent in 2023. A significant portion of edge customers (51%) have small deployments (two to three servers), with a notable percentage (34%) operating on freemium models. However, there is a growing momentum with 24% of edge customers already with small deployments across multiple data centres.

Figure 1: To what extent are customers deploying multiple network edge data centres?

Source: STL Partners network edge survey, August 2023

Figure 2: To what extent are customers paying for network edge?

Source: STL Partners networke edge survey, August 2023

Most telcos have not taken edge deep in their networks yet, typically one node per urban area with over two million population. While this is sufficient for market testing and meeting current demands, it is evident that telcos have yet to fully scale their edge computing deployments. This trend spans across various markets for example, although Verizon’s Wavelength deployments may appear more extensive than Vodafone’s (with 19 nodes in the US in comparison to Vodafone’s 6 across its European footprint), the difference aligns with the higher concentration of urban areas with over two million inhabitants in the United States.

Deployments completed in partnership with hyperscalers tend to be less deeply distributed compared to those undertaken solely by telcos themselves. For instance, in the United States, Lumen has about the same number of edge computing sites as there are urban areas with over one million population. In contrast, Verizon’s collaboration with AWS results in less than half the number of sites, reflecting a closer alignment with urban areas with double the population density.

Figure 3: Telcos are yet to scale network edge deployments

Source: citypopulation.de, STL Partners

See how STL can help you capitalise on the edge computing opportunity

Develop a rich understanding of the edge computing opportunity with our research service. Book your demo today

Book a demo

Analysis of lessons from early movers

From our research, we identified four key lessons that the industry should take from the experience of telco network edge early movers.

Key lesson 1: The value proposition for telco edge should be articulated beyond low latency

Today, early movers are directing their focus towards data localisation and bandwidth management, or they are simply presenting their offerings as a private cloud operated by a local telecoms operator. During our discussion regarding telco edge deployments, Lumen emphasized that latency is not necessarily the primary value proposition for telcos. Few workloads require 5-20ms latency, and for those that do, on-premise solutions may be preferable for their resilience. Service Level Agreements (SLAs), security measures, tools, and data costs often hold equal or greater importance from a value proposition perspective.

Key lesson 2: Expect relatively low edge infrastructure and platform ROI and look beyond simply emulating cloud services

The industry should view edge computing as a long-term growth opportunity rather than a means for short-term returns, and allocate investments accordingly. Substantial upfront investment is necessary for launching either home grown or hyperscaler edge solutions, involving costs such as data centre upgrades and equipment procurement. Lumen serves as an example of a telco experiencing a relatively prolonged payback period from its initial distributed edge node investments. While demand is currently sufficient, a more gradual deployment in response to customer demand would have been preferable.

Furthermore, achieving optimal capacity deployment is challenging and costly. Lumen, for instance, has encountered difficulties in ensuring sufficient capacity in the right place while avoiding low utilisation. This challenge intensifies as more sites and Service Key Updates (SKUs) are offered, although disaggregated composable infrastructure can help.

Consequently, it is crucial for telcos to understand the long-term vision, set realistic expectations, and foster internal commitment to the strategic rationale for edge computing. Maintaining clear organisational ownership, corporate patience, and goals beyond immediate revenue attribution are essential for long-term success.

Key lesson 3: Telcos should focus on use cases and integrated solutions where mobile network edge is an enabler

During our discussion, SK Telecom highlighted the mistake of solely focusing on the edge platform and expecting customers to come and deploy on it. Instead, telcos need to develop a robust edge services strategy as customers may not be aware of the services they require. Telcos should consider developing solutions independently or collaborating with specialised providers to deliver integrated solutions using edge technology, such as analytics and artificial intelligence (AI). SK Telecom sees AI as the primary driver for edge computing and is focusing its marketing efforts on “AI near to customers.”

Key lesson 4: Telco edge business cases should assume investment in building their own telco edge and traffic flow optimisation

While hyperscaler partnerships have been beneficial for many early movers in network edge, there are notable limitations with this approach. SK Telecom highlighted that AWS offers generic computing at scale, serving only developers and enterprises that already use AWS Cloud. Wavelength just extends the AWS cloud footprint but lacks support for enterprises needing advanced customisations or specialised chipsets.

In contrast, deploying telco’s own edge infrastructure allows for differentiation and value addition through customisation. For instance, SK Telecom’s current focus is on reducing the power consumption of its edge nodes by implementing less power-intensive chipsets.

Telco-offered services should delve deeper to meet the growing demands for traffic and latency requirements. The evolution of orchestration platforms, network APIs, and traffic routing in recent years has facilitated the realisation of telco edge capabilities. Therefore, telcos need to invest in their own edge nodes to ensure long-term viability and competitiveness.

Conclusion

The insights learnt from early telco edge movers provide valuable implications for the telecoms operators for their future deployments in network edge computing. Understanding the evolving market dynamics, investing strategically, and prioritising use case development are important for telcos to capitalise on the edge computing opportunity. This article only discusses part of our findings from our survey and interviews with the five telco early movers. For a comprehensive analysis of their edge footprint and strategic implications, please access the full report here.

Zoey Yin

Zoey Yin

Consultant

Zoey Yin is a Consultant at STL Partners, specialising in edge computing.