The edge refers to endpoint devices used by people or
embedded in the world around us. Edge computing describes a computing topology in which information
processing, and content collection and delivery, are placed closer to these
endpoints. It tries to keep the traffic and processing local, with the goal
being to reduce traffic and latency.
In the near
term, edge is being driven by IoT and the need keep the processing close to the
end rather than on a centralized cloud server. However, rather than create a
new architecture, cloud computing and edge computing will evolve as
complementary models with cloud services being managed as a centralized service
executing, not only on centralized servers, but in distributed servers
on-premises and on the edge devices themselves.
Over the next five years, specialized AI chips, along with
greater processing power, storage and other advanced capabilities, will be
added to a wider array of edge devices. The extreme heterogeneity of this
embedded IoT world and the long life cycles of assets such as industrial
systems will create significant management challenges. Longer term, as
5G matures, the expanding edge computing environment
will have more robust communication back to centralized services. 5G provides lower
latency, higher bandwidth, and (very importantly for edge) a dramatic increase
in the number of nodes (edge endoints) per square km.