There is no definitive location or range of locations that edge computing must exist — nodes can include network routers, cell or radio towers, WiFi hot spots, DSL-boxes, and local data centers. Edge Computing is the concept of placing computing power near the user’s location with the primary goal of reducing latency. Edge Computing nodes are on the outer region of the core network or its backbone. Almost any device with computational power that is near or at the user’s location can act as an edge computing device, as long as it’s practical.
Typical Edge Locations: Enterprise and Service Providers
Edge locations can include cell towers, on-premises servers, and small network devices. Cell towers are attractive edge computing locations because they have space to place small data centers with direct access to the tower’s existing network resources and power. In addition, these towers are often in high-service areas.
IoT solutions company IntelZONE said in a post that gateway devices like switches or routers also can be edge computing devices if the machine processes and analyzes data with its excess resources. A report by Business Insider Intelligence included industrial PCs and micro data centers as edge locations.
Industrial Edge Locations
Industrial IoT (IIoT) use cases put computation power on nearly any connected device. The Industrial Internet Consortium (IIC) published a white paper titled “”Introduction to Edge Computing in IIoT in 2018. It describes how edge computing is used in an industrial setting: Data analysis happens on a machine’s sensors or on wearable devices so that people monitoring the devices can make the appropriate responses with as little delay as possible. Pertinent data that is derived from the analysis is then sent to the larger local on-site network and then to the central cloud.
The paper gives examples of edge computing devices in various use cases:
- Personal vital sign monitors for crew safety management
- GPS trackers for vehicle fleet tracking
- Infrared sensors for predictive maintenance
The types of edge computing deployments and boundaries present in IIoT. Source: IIC
The Changing Definition of a Data Center
Edge Computing is a form of a distributed cloud. The centralized cloud and its associated data centers are almost always further from the user than ideal for applications that require low latency. However, data centers that are not part of the central cloud do exist, and can be closer to a user. Such a regional data center may not be closer than another network node, but it would still be considered an edge computing resource because it is not part of the core network.
There are also products (like Dell EMC’s micro data center) that enterprises can purchase to place on premises. This is indicative of the evolving nature of data centers: they no longer have to be so central. In a 2016 white paper by Schneider Electric titled “”The Drivers and Benefits of Edge Computing, so-called localized data centers that are just one to 10 racks of servers on the premises of an enterprise are an example of an edge computing device. That paper also details the above-mentioned regional data centers.
In a 2019 SDxCentral eBrief titled “”The Impact of Edge Computing on the Data Center, Mike Capuano, CMO at Pluribus, said, “An edge data center is really anything other than a centralized data center.” With such a broad and blurry definition, edge has become an umbrella term for any network node outside of the core network.