In this lecture, we will look at edge computing and fog computing
Fog computing is a decentralized architecture where we take cloud capabilities to the edge of the network
Since there will be multiple layers of fog, the data processing of fog will be in multiple layers between cloud and edge devices
The focus of edge computing is that the data will be processed on the source
If we want to minimize the latency of the device and check the distance of the data, then we have to minimize the minimal distance of the data
The data processing in fog computing is on fog nodes which can be anywhere between cloud and edge
This basically allows distributed processing
In edge computing, the data is processed directly on the device or very close to it in gateways
In this, we basically focus on immediate processing with minimal delay
Now, the point of scalability is that fog computing allows you to use multiple fog nodes in the network
and large volumes of data can be effectively dealt with with numerous sources
In edge computing, scalability is an issue because there are individual edge devices and their capability is limited
So, there is a capability of processing requirement, so scalability can be a problem in terms of processing
If we look at the point of resource management, fog computing utilizes the network resources and uses them in connectivity and communication
And it also has to balance the fog nodes and centralized cloud resources
In edge computing, there is a dedicated hardware to process tasks, which limits flexibility
And every individual edge device is optimized for resource
If we look at the point of security considerations, fog computing gives enhanced security
because multiple layers of management and control exist
and comprehensive security protocols can be implemented across nodes
In edge devices, security is dependent on the individual device
So, if you want to secure more, then additional protection will have to be given to you
Because the number of devices will be high and if you want to secure more, then the edge computing devices will be vulnerable
So, from the point of interoperability, fog computing can interoperate multiple device types
and allow seamless communication
While in edge computing, executing tasks are independent, you can do it without interacting with anyone
Now, the problem is that there are multiple devices and diverse systems, so how can they be integrated?
So, this problem can arise in edge computing
If we look at the use cases, fog computing is a very common example of smart cities
So, excel in scenario is requiring large scale deployment for fog computing
Edge computing will perform better when there is a smaller volume of local processing tasks
That is, there is local processing and small volumes of data
So, in autonomous vehicles, edge computing can be a good use case