What Does the Growing Popularity of the Edge and the Cloud Hold for Enterprise 5G?

Enterprise cellular connectivity within enterprise verticals remains in a somewhat complicated situation. While Communication Service Providers (CSPs) and infrastructure vendors originally positioned enterprise 5G to be able to address both mission-critical and non-mission-critical use cases within the same infrastructure, it has now become clear that despite 5G being available today, the important enterprise 5G capabilities—mainly Ultra-Reliable Low Latency Communication (URLLC) and full support of Time-Sensitive Networking (TSN)—are still not available in the market.

While both telco infrastructure vendors and CSPs, by now, have realized that they have overpromised and underdelivered when it comes to enterprise 5G, they only have limited influence over how fast commercially available 5G infrastructure will catch up and deliver what the telco industry kept promising to enterprises.

The main inhibitor for deployment is the ongoing bottleneck of availability of industrial-grade devices, particularly for 5G. Most of the important features for industrial enterprises (i.e., URLLC, full support for TSN, or Massive Machine-Type Communication (mMTC)), require 5G capabilities that have been standardized in The 3rd Generation Partnership Project’s (3GPP) Release 16. Despite this release being frozen (which is telco language for completed) in the summer of 2020, industrial-grade devices compatible with this Release are expected to emerge only by the end of 2023 and reach scale in 2024.

At the same time, further digitization of workflows and processes will inevitably increase the amount of data being generated on factory floors, within shipping facilities or airports, and other enterprise environments. This, in turn, will increase the demand for computing within enterprises and will, therefore, attract interest of new players, including hyperscalers and computing companies. The growing presence of Amazon Web Services (AWS), Microsoft, and Google Cloud at industrial trade fairs (e.g., Hannover Messe or the International Manufacturing Technology Show (IMTS), where AWS was present for the first time this year), is an important indicator of this change. To make sense of these data, however, it will need to be stored, analyzed, and processed in both an effective and efficient way. Two factors will determine the role of the cloud and the edge in this context.

1. Total Cost of Ownership (TCO): Enterprises will strive to minimize their TCO of any digitization solution. As ABI Research discovered in countless discussions with enterprises, TCO is the major Decision Forming Factor (DFF). The drive for cost reduction for enterprise digitization solutions will raise questions of how existing, public infrastructure can be used most efficiently. Using a hybrid cloud setup can deliver important TCO savings for enterprises—the industry talks about savings of up to 50%. In addition, enterprises (especially Small and Medium Enterprises (SMEs)) are short on networking experts and will, therefore, be looking outsource as many network control functions as possible. As a result, public infrastructure will remain interesting to enterprises—with the limitations discussed in the next paragraph.
2. Network Data Integrity: Enterprises will continue to demand a heightened degree of network integrity. This could stem from either intrinsic or extrinsic motivation. Industrial environments, for example, often deal with highly-sensitive data around the number and condition of their production assets, or their security mechanisms. All of this is highly-sensitive data that enterprise owners do not want to see leaving their premises, exposing them to attacks or unauthorized access from competitors. Therefore, network integrity, in this case, is intrinsically motivated. Other verticals, like healthcare for example, are governed by a strict regulatory framework that requires patient data to remain within the premises of a hospital or doctor’s office. In these cases, the drive for network integrity is extrinsically motivated. Regardless of whether intrinsically or extrinsically motivated, enterprise deployments will continue to require on-premises edge deployment for data storage, processing, and analysis.

The changing dynamics of the edge and the cloud and their prominence within enterprise verticals also changes the realistic role that connectivity will play within digitized enterprises. Certain enterprises will use cellular connectivity to connect their production assets with each other in an Internet of Things (IoT) type of scenario, but as the introduction to ABI Insight argues, these deployments have failed, so far, to gain any noticeable scale.

The growing importance of edge and cloud deployments in the enterprise context opens a new application scenario for connectivity technologies for enterprise digitization projects. Enterprises will require technology to reliably connect the on-premises edge compute units with any public and/or private cloud. Admittedly, these back end use cases are not as shiny and fit for marketing as a fully remote surgery, lights-out factory, or remote tattoo painting session, but they will be much more realistic, as the edge and the cloud are already becoming integral parts of enterprise digitization. Furthermore, concentrating on connecting edge and cloud computing with each other does not necessarily require the same features as a 5G private network deployment for IoT use cases (i.e., URLLC or TSN) and would, therefore, avoid the bottleneck of missing Release 16-capable industrial-grade devices. To benefit from this trend, the telecoms industry needs to move closer toward the respective end verticals, with edge computing providers to equip edge computing units and servers. Combining private cellular connectivity with edge computing in a single product, ideally offered in an as-a-Service business model, is expected to resonate particularly well with end markets.

While large, multinational corporations might be willing to multisource components, SMEs will look for a single edge solution provider that will have the necessary connectivity element already implemented. This, in turn, will require computing companies to open up toward infrastructure vendors and chipset manufacturers to integrate cellular connectivity into their offerings. From an organizational point of view, this could be done by either co-creation initiatives or robust partnerships. From a technology solution point of view, CSPs could also apply a Network-as-a-Service (NaaS) business model to their private network solution. By using Open Application Programming Interfaces (APIs) and applying these to the NaaS platform, third parties can integrate services into the platform. This would not only drive operational and business value for the SMEs, but it would also open up a second revenue stream for CSPs.