Edge computing is not just a methodology but also a philosophy of networking that is primarily focused on bringing computing devices closer to the network. The objective is to reduce any sort of latency in the usage of bandwidth. To put it in layman’s terms, edge computing means executing a smaller number of processes in the cloud and migrating those processes to a more localized environment such as a user’s computer, an IoT device, or an edge server. Executing this process ensures a reduction in long-distance communication that arises between the client and the server.
For all internet devices, a network edge is a place where the device or the local network that contains the device, communicates with the internet. One can call the word edge a buzzword and its interpretation is rather funny. For instance, the computer of a user or the processor inside an IoT device can be treated as a network edge device; however, the router used by the user, or the ISP is also factored as a network edge device. The point to be noted here is that edge of any network, from a proximity point of view is very close to the device; unlike other scenarios involving cloud servers.
Difference between edge computing and other computing models
Historically speaking, early days computers were large, bulky machines that could be accessed either via a terminal or directly. However, with the invention of personal computers, which was quite a dominant computing device for quite a long time, the methodology of computing was more in a distributed manner. Multiple applications were executed, and the data was either stored in the local computer or probably stored in an on-premise data center.
However, with cloud computing, we are seeing a paradigm shift in the way the computing process is done. It brings a significant value proposition where data is stored in a vendor-managed cloud data center or a collection of multiple data centers. Using cloud computing technology, users can access data from any part of the world, through the internet.
But the flip side is that because of the distance between the user and the server location, the question of latency may arise. Edge computing brings the user closer to the server location, making sure that the data does not have to travel a distance. In a nutshell
- The early days of computing involved applications running in a single computer and data also stored there
- Personal Computing that resulted in decentralized applications that were operating in a local environment
- Cloud computing involves applications running centrally in data centers
- Edge Computing ensures that the applications are closer to the users and data is stored either in the local device or in the edge server.
Examples of edge computing
Let us consider a situation where there is a building that has multiple high-definition IoT sensor cameras. These cameras just provide raw video footage, and they consistently stream the videos to a cloud server. On the server, the videos undergo processing through a motion detection application that captures all movements and stores the video footage in the cloud server. Imagine the amount of stress that the building’s internet infrastructure undergoes because of the high consumption of bandwidth due to heavy video footage files. On top of this, there is a heavy load on the cloud server as it has to store these video files.
Now, if we move the motion sensor application to the network edge, each camera can harness the power of its internal computer to run the motion sensor application and then push it to the cloud server as and when needed. This will bring about a considerable amount of reduction in bandwidth usage because a major chunk of camera footage won’t be required to travel to the cloud server.
Furthermore, the cloud server will now be storing only the critical video footage, unlike the entire dump in the previous case.
Potential use cases of edge computing
- Monitoring of security systems as mentioned above
- Smart IoT devices that can connect to the internet by running the application or code within the device itself instead of doing it on a cloud server
- Self-driven cars that need to have instant reaction instead of fetching information from the server
- Medical devices used for monitoring critical parameters need to operate on a real-time basis instead of waiting for updates from a server
|Advantages of edge computing||Disadvantages of edge computing|
|Eliminates latency resulting in better performance||Increase the possibility of attack vectors questioning the security of the infrastructure|
|Reduction in bandwidth that brings about great cost savings||Requirement of more local hardware causing a question on maintenance of the system|
|Eliminates congestion arising due to usage of a large volume of data||The cost of implementing edge computing can be very expensive.|
|The process of implementation of edge infrastructure is extremely complex||Edge computing can only process a limited set of data. The data to be processed|
The implementation and adoption of edge computing have brought about a paradigm shift in the domain of data analytics to a new dimension. More organizations are dependent upon this technology, which is completely data-driven, and organizations that require instant and lightning-fast results. There are many online platforms that provide certified courses on edge computing.
It does not matter what kind of edge computing is of interest to you – be it cloud edge, IoT edge, or mobile edge, it is important that the right solution can help in achieving the following organizational goals:
- Manage large scale distribution of software
- Harness the power and flexibility of open-source technology
- Tie up with a partner who is trusted in this space and has the right domain expertise
- Manage any concerns around the security of edge computing infrastructure