This blog serves as a reference point to Edge Computing in 2021. Lately, business investments and research intrigue into Edge Computing have escalated drastically. Computing at the Edge shifts computing and storage away from the cloud to the edge of the internet in close quarters to mobile devices and sensors. 

Therefore, Edge computing makes is viable to provide very responsive services for mobile computing, scalability, and privacy-policy enforcement of Big Data-based AI processing at the Edge; otherwise known as AIEdge. 

In this blog, we will give you a simple understanding of a few concepts in the form of a guide. These concepts are: 

• What Edge Computing is
• Where it came from, and the benefits and advantages that it provides
• Connected concepts, edge vs fog computing 

Defining Edge Computing 

Edge computing is the new kid on the block. It is a paradigm in computing that executes computing at the ‘edge’ of networks. In contrast to cloud computing, edge computing shifts nearer to the user and is in closer proximity to the origin of the data/information. On the edge of the network, it is lightweight for local and smaller scale information storage and process. 

With the quickening pace with which the Internet of Everything (IoE) is evolving, the instances of smart devices that are on the internet is escalating very quickly, producing large-scale data at the network edge – which you may otherwise know as ‘Big Data’. This enormous amount creates issues like bandwidth load, reduced response speeds, weakened security, and reduced privacy in conventional cloud computing models. 

Therefore, conventional cloud computing is not adequate to assist the requirements of today’s specifications for smart services and smart data processing, therefore, there is a pressing need that is created for Edge Computing technologies. 

Edge computing diverts computational information, apps, and services away from traditional cloud servers to the edge of a network. App developers can utilize the on-device computing systems by providing the users services nearer to them. Edge computing is identified by its high bandwidth, ultra-low latency, and real-time access to network data that can be utilized by various apps. 

Hence, Edge computing is at the base of next-gen Edge intelligence, the rollout of ML algorithms to the edge device where the information is produced. 

A deeper dive 

As Edge computing has become a hotbed for research and innovation, there are various technical definitions that dissect the meaning of Edge computing even further. Academicians at Carnegie Mellon University have stated that Edge Computing is “a fresh computing model that places computing and storage assets like cloudlets, micro data centers, fog nodes, etc., at the edge of networks, in closer proximity to mobile devices or sensors”. 

China’s Edge Computing Consortium states that Edge Computing as “close to the edge of the network or the origin of the information/data, an open platform which features integration of core capacities like computing, networking, storage, apps, and furnishes edge intelligent services in close proximity attain the industry agility key requirements with regards to connection, real-time business, security and privacy, app intelligence, and information optimization. 

Zha gives this definition for Edge Computing: “Edge computing is a fresh computing model that brings about a union of resources that are in close proximity to the user in geographical distance, or network distance to give network, storage, and computing for applications service.  

Some history and the need for Edge Computing 

2021’s smart society is compelled by the necessity for smart and connected services and products across several industries. Edge devices have permeated to several pockets of society, like smart homes, security cameras, self-driving vehicles, smart production robots, and a lot more. The quantity of devices that are connected is persistently increasing.  

Hence, the vast data output produced by devices and gadgets globally is dramatically escalating. On the basis of the ongoing and enormous growth of information volume and several data processing requirements, cloud-based big data processing has demonstrated various drawbacks:

1. 1. Real-time: Transmitting huge amounts of information to cloud creates the outcome of a big load of network transmission bandwidth, causing delays in transmitting information. Cloud computing is not equipped to fulfill these real-time business requirements.
   2. Power consumption: The energy usage of data centers has escalated considerably. Cloud computing has proven unable to fulfill the escalating demand for optimization of energy consumption.
   3. Privacy/security: Upload of information to the cloud and recording them in a centralized environment carries with it the drawbacks of privacy leakage or security compromises in the form of attacks. The possibility of information en-route attacks or hijacks is larger in cloud computing than in edge computing due to the longer route to the server. 

Hence, the objective of Edge Computing is to furnish services and execute computations at the edge of the network and data production. The objective of Edge computing is to shift the cloud’s network, computing, storage capacities, and resources to the edge of the network, and furnish smart services at this edge. 

This is needed to fulfill the crucial requirements of the IT sector in real-time business, app intelligence, privacy/security, information optimization, and to fulfill the necessities of low latency and increased bandwidth on the network. 

References 

https://viso.ai/deep-learning/edge-computing-a-practical-overview/