Before looking into types of edge computing you should know what is edge computing and why do we need that. There have been plenty of reasons why organizations can start employing edge computing technologies in their infrastructure. From real-time AI data analysis and improved user experience, to dramatically lower running costs and planned downtimes.
Types of Edge Computing
What edge computing is basically, seems ambiguous and varies depending on who you ask. This post focuses on types of Edge Computing. There are four types of edge computing.
Cloud:
The first type of edge computing is Cloud. It refers mainly to large data centers run by cloud companies such as AWS, Azure, and GCP. Which may include VMware Cloud on AWS and other cloud or provide for too. The cloud’s key attributes centralized and runs at scale. The downside is you have very high availability of infrastructure. And also, access to a lot of facilities, and an unlimited amount of money.
The downside clustered. There is no guarantee of network access to sensors or computers at the edge and high latency. Internet traffic to and from the cloud also most likely entails an expense.
Device Edge:
The second type of edge computing is the Device edge. This consists of one or more tiny servers and also known as a nano DC. It would consist only of one or a few customize and would have limited processing power. Databases in this segment would likely not mounted on a rack. And also, we would need to be able to run without refrigerating.
These are also located in places that are not typically associates in data centers. Such as warehouses, wind generators, and durable to cope with harsh weather. They positioned right next to IoT sensors, so there limit issues in latency, bandwidth, or communication. The downside is that these small devices can only have low power and facilities.
Compute Edge:
The third type of edge computing is the Compute edge. It is also a micro-DC. Anyway, it is a small data center consisting of everything from a few up to many server racks. They are usually located near or next to IoT devices, and may also be needed for reasons of local enforcement. The point is that these server farms have ventilation and so on, and have regular servers installed on racks.
In these data centers, you will have quite a lot of resources. If, not the same variety of facilities and capabilities as in the cloud. One advantage is that system latency at the edge would be less relative to the cloud. So network bandwidth should be higher but more efficient.
Sensor:
The final type of edge computing is the sensor. IoT sensors are instruments that either collect data or monitor things. Some sensors around the world like a surveillance camera, clock, light bulb, etc. Depending on the bandwidth, latency, or communication requirements, they would usually not include any computing power in themselves. And also, it will instead communicate with the system edge, network edge, or cloud.
Conclusion:
The hope is that potential systems will spread through these different layers of computing, communicating with devices at the bottom. Some parts of the application would require direct exposure to edge sensors. It would require to be in the System Edge. Whereas, other parts would need exposure to more complicated resources. And also more processing power in either Network Edge or the Cloud. Components at different locations will work together to shape an application.
All you need to know about IoT
| Introduction to IoT | Career Options after IoT |
| IoT in Business Growth | How IoT works? |
| Future of IoT | Benefits of IoT |
| Disadvantages of IoT | Salary After IoT Course |
Learn IoT
| Top 7 IoT University/ Colleges in India | Top 7 Training Institutes of IoT |
| Top 7 Online IoT Training Programs | Top 7 Certification Courses of IoT |
Learn IoT with WAC
Other Skills in Demand
| Artificial Intelligence | Data Science |
| Digital Marketing | Business Analytics |
| Big Data | Internet of Things |
| Python Programming | Robotics & Embedded System |
| Android App Development | Machine Learning |
