Key Criteria to Move Cloud Workloads to the Edge
EDGE NETWORK ARCHITECTURE Figure 1 is an overview of edge network architecture .
EAST-WEST COMMUNICATION
DATA CENTER
NORTH-SOUTH COMMUNICATION
PRIVATE PUBLIC HYBRID
EDGE NODE
EDGE NODE EDGE NODE EDGE NODE
LAYER
LAYER
EDGE NODE EDGE NODE
IoT DEVICES |
IoT DEVICES |
|
|
IoT DEVICES |
IoT DEVICES |
Fig . 1 - Industrial Internet Consortium Distributed Computing in the Edge Framework .
There is a hierarchy of computation , storage , and networking resources in this architecture . The data center implements the cloud . Below that are multiple layers of edge computing . Finally , the IoT devices shown can themselves implement some computation and storage capabilities ( for example , an intelligent security camera with built-in video analytics capability ). North-south communications links provide interconnect between the layers . East-west communications links interconnect peer devices on the same layer .
The focus of this article is to understand why certain workloads cannot be optimally performed on the cloud or the intelligent IoT devices and must ( at least partially ) be moved to one or more layers of edge nodes . It discusses the most important selection criteria when deciding to move computational workloads from cloud data centers to edge computing nodes .
IIC Journal of Innovation - 3 -