Edge server shipments set to double by 2024

With the internet of things (IoT) and campus computing rising rapidly, latency and bandwidth are becoming key performance determinants, driving the need for better comms networks and more computing power to be placed closer to end-users and machines – and the net result will be that the number of servers deployed at edge locations will double over the next five years, predicts analyst Omdia.

Further driving the market will be the continued reliance of enterprises and consumers on cloud services for remote working, learning, social interaction and entertainment.

Omdia defines the edge as locations with a maximum 20 milliseconds (ms) round trip time (RTT) to the end-user, device or machine. This includes telco IT-operated sites such as central offices and regional datacentres, cloud service provider points of presence with equipment deployed at colocation provider datacentres and enterprises deploying servers at branch offices, retail stores, campus sites and colocation provider datacentres.

Using these parameters, the analyst calculates in its research that a total of 4.7 million servers shipped in 2024 will be located at the edge, and almost one-third (32.2%) of servers shipped to telcos in 2024 will be deployed at edge locations, up from 22.8% in 2019.

It notes that a large portion of the servers that telcos have deployed at the edge so far are used for content delivery networks (CDNs) and that surveying telcos, many edge located server deployments will be justified by the cost savings achieved by virtualising network functions, followed by revenue-bearing new services.

“The development of new devices and software technologies in response to growing requirements to improve business processes, relieve humans of repeatable tasks, and make life more fun is accelerating global computing demand,” said Vlad Galabov, principal analyst for datacentre IT at Omdia. “At the same time, the nature of devices and applications is changing, with the collection and real-time processing of data becoming increasingly important.

“By 2024, a smaller portion of telco edge-located servers will be running CDNs or virtual network functions [VNFs] in favour of other workloads, such as video content delivery, vehicular communication for autonomous vehicles, gaming and augmented/virtual reality [AR/VR].”

The key driver for these edge deployments will be the massive uptake of cloud services, said Omdia. It expects that, by 2024, 12.2% of servers shipped to hyperscale cloud service providers (CSPs) will be deployed at edge locations, up from 5% in 2019.

Drivers for the growth of the hyperscale CSP edge include the popularisation of video streaming, which requires effective content delivery networks, and the pursuit of new services such as cloud gaming, PC as a service, and cloud AR and VR, where the latency between the end-user inputting commands and the cloud-located server becomes critical. 
 

The study also revealed that tier-two CSP edge will represent 36.7% of all servers shipped to tier-two CSPs in 2024, up from 31.4% in 2019. Many of the local second-tier CSPs, which largely operate within a country, will, by default, have their datacentres located within very close proximity to their end-users.

Omdia sees this close proximity as a unique differentiator for local CSPs when faced with competition from the hyperscalers. The firm’s latency tests between servers located in the same region/city as end-users showed that nearly half of those servers had sub-20ms RTT.

The forecast also took into account the local tier-two CSPs’ close proximity to the end-user/device and noted the fact that larger tier 2 CSPs such as IBM, Apple, Oracle and SAP currently have a more centralised server deployment model, resembling a hyperscale CSP. Omdia expects the latter to increase their number of edge locations to launch new services in an attempt to compete with hyperscale CSPs. 

Omdia’s research also found that enterprises were early adopters of edge computing for latency sensitive workloads, such as healthcare and industrial applications; on-site data consolidation, data sharing and analytics; and retail store management.  

“One driver for enterprises moving more servers to the edge is the expansion of automated manufacturing and the use of IoT devices,” said Galabov. “Just a few examples of latency-bound workloads that spring to mind are smart factory motion control, real-time plant asset management and self-diagnosis, augmented reality for factory management and repair, robot-aided surgery and new retail store operation models.”