Figure 13-6 Community Cloud
Edge and Fog Computing
With the increase in IoT applications, devices, and
sensors, the various deployment models covered so far
are just not up to the task of handling the sheer volume
of data being created and needing to be processed. Self-
driving cars, robotics, computer video processing, and
many other use cases can produce gigabytes of data in
real time. This sea of data cannot be consumed and
processed by a centralized cloud application, given the
cost of transporting the data and the latency involved in
receiving information back to do anything with it. Your
self-driving car would run into a ditch before a cloud
service could detect and respond.
These constraints required engineers to think differently
about use cases requiring intelligence at the edge. A new
class of application deployment was created, called fog
computing (because fog is basically a cloud on the
ground).The idea behind fog is to leverage the scale and
capabilities of cloud models to handle the control and
deployment of applications at the edge. With edge
computing, you let local processing of data occur closer
to the actual sensors and devices where it can be acted on
more quickly. Think of fog as a framework and structure
that enables edge computing. Figure 13-7 shows how
edge and fog work together.