Title: Review on Edge Computing Author: K M V Ram Gopal (Student ID: 700733717) 1. INTRODUCTION Innovation got a tremendous distinction our lives. With the expansion in innovation use of the Internet has been a piece of living souls with the advancement of insightful society and the nonstop improvement of individuals' necessities, knowledge has involved different ventures and individuals' regular routines in the public eye. Edge gadgets have spread to all parts of society, like brilliant homes and independent vehicles in the field of transportation, camera, smart creation robot in keen assembling, and so forth subsequently, the quantity of gadgets associated with the Internet has expanded altogether. Cisco called attention to in the Global Cloud Index [1] that in 2016, there were 17.1 billion gadgets associated with the Internet, by 2019, the all-out number of information traffic in worldwide server farms will arrive at 10.4 Zettabyte (ZB), 45% of the information will be put away, handled and investigated on the edge of the organization, and by 2020, the quantity of remote gadgets associated with the organization will surpass 50 billion. How much information created by gadgets worldwide has additionally expanded from 218ZB in 2016 to 847 ZB in 2021 worldwide information organization Internet Data Centre (IDC) measurements show that by 2020, the quantity of terminals and gadgets associated with the organization will surpass 50 billion, and the complete worldwide information in 2020 will likewise surpass 40 ZB [2]. Cloud based short-comings: I. Real time ii. Security and privacy iii. Energy consumption 2. EDGE COMPUTING a) Edge computing concepts: Edge processing is not the same as conventional distributed computing. It is another processing worldview that performs registering at the edge of the organization. Its centre thought is to make figuring nearer to the wellspring of the information [3]. Analysts have various meanings of edge registering. Shi et al. [4]-[6] presented the rise of the idea of edge figuring: ''Edge processing is another registering method of organization edge execution. The downlink information of edge figuring addresses cloud administration, the uplink information addresses the Internet of Everything, and the edge of edge registering alludes to the inconsistent processing and organization assets between the information source and the way of distributed computing place.'' Satyanarayanan, a teacher at Carnegie Mellon University in the United States, depicts edge processing as: ''Edge figuring is another processing model that sends registering and capacity assets, (for example, cloudlets, miniature server farms, or haze hubs, and so on) at the edge of the organization nearer to cell phones or sensors'' [4]. Zha et al. [8] proposed based on the over two definitions: ''Edge figuring is another registering model that brings together assets that are near the client in topographical distance or organization distance to give processing, stockpiling, and organization for applications administration.'' b) Edge computing and cloud computing: The rise of edge figuring won't supplant distributed computing. In the parts of the organization, business, application, and knowledge, the two should exist together, complete one another and create in a planned manner, which will help the advanced change of the business indeed. All information onto edge hubs actually should be summed up in the cloud to accomplish top to bottom investigation and get more significant examination results Thusly, distributed computing is as yet playing a significant job in the improvement of Internet of Things gadgets that are slowly wise. With regards to the Internet of Things, on the off chance that all the huge measures of information created by the associated gadgets are communicated to the cloud, distributed computing will cause a huge burden. Right now, edge figuring is expected to share the tension of the cloud and assume responsibility for errands inside the extent of the edge. At the point when there is an issue in edge registering, the information in the cloud isn't lost. c) Advantages of edge computing: Edge computing model stores and processes data on edge devices without uploading to cloud computing platform. Application situation Network bandwidth pressure Real-time Calculation mode Cloud computing Global More High Large-scale centralized processing Edge computing local Less Low Small-scale intelligent analysis 3) ARCHITECTURE: Edge computing is one of the important technologies in the future generation of communication networks, after the Internet of Things and artificial intelligence [25], with the Internet of Everything period and the development of 5G. Many firms are focusing on the edge computing reference architecture. This section starts with a broad overview of edge computing architecture, then dives into the reference design suggested by the edge computing industry association (ECC) and the Linux Foundation. General architecture: Edge computing architecture is a federated network topology that introduces edge devices between terminal devices and cloud computing to extend cloud services to the network's edge. The structure of cloud-edge collaboration is generally divided into i) Terminal layer ii) Edge layer iii) Cloud computing layer b) EDGE COMPUTING REFERENCE FRAME 3.0: Model-driven engineering is used to create the frame of reference. We must achieve the following four aims in order to represent knowledge of the physical and digital worlds: 1) Create a real-time and systematic cognitive model of the physical environment and accomplish physical-digital collaboration; 2) Create a reusable knowledge model system in each vertical industry based on modelling approach and complete cross-industry ecological cooperation 3) System-to-system, service-to-service, and other model-based interfaces for interaction, in order to decouple the software interface from the programming language and reduce system heterogeneity. 4) Can effectively support the development service's life cycle, including deployment, data processing, and security. The architecture of the edge computing reference architecture is depicted in this diagram. 4) APPLICATIONS: i) Edge Computing Video Cache: Popular TV shows, movies, and other video resources with high download frequency are cached on the local MEC server using the edge computing intelligent analysis function (based on search heat). When a user requests that a video be played, the video resource can achieve the effect of loading from the local, conserving bandwidth while also drastically lowering the user's waiting time. Furthermore, content in the MEC platform is optimized based on RAN-side perception, allowing content to be dynamically optimized based on network real-time information (network load, link quality, data throughput rate, and so on), resulting in an improvement in Quality of Experience (QoE) and network efficiency. ii) Edge Calculation And 5g: 5G provides the advantages of low latency, high bandwidth, and high capacity, which eliminates many of the difficulties that exist in traditional communication, but also leads to rapid data volume increase. It is critical to create a solid, relevant, and executable business model at this point. The properties of 5G, such as fast processing and low latency, can give a new means for quick response and can optimize the end, edge, and cloud all at the same time. As a result, the advancement of edge computing technology is directly linked to the development of 5G: on the one hand, edge computing can support 5G, and on the other hand, edge computing is a key component of 5G. Edge calculation, on the other hand, can be implemented in a variety of ways because 5G is expressed as software. iii) Edge Computing Network Video Live Broadcast: The network video live broadcast system is a multimedia network platform that seeks to broadcast live audio and video live events such as current competitions, conferences, concerts, and instruction to remote audiences in real time. Despite the fact that the server port of a regular live video broadcasting system typically uses 100 megabytes or gigabytes of network bandwidth, the latency problem in the entire process cannot be overlooked due to the enormous audio and video files. Edge computing is used to overcome this problem in this field. iv) Predictive Maintenance: Preventive maintenance is currently used by the majority of enterprises to increase the stability of their production processes. Preventive maintenance is typically done on a regular basis. Regular maintenance reduces the likelihood of equipment failure or shutdown in a short period of time. Preventive maintenance can reduce overall production line downtime and the amount of failures caused by equipment deterioration, and it has the benefits of being simple to install and operate. However, because maintenance time is influenced by experience, insufficient or excessive maintenance may occur. v) Security Monitoring: One of the most essential methods for humans to learn about the world and receive knowledge is through vision. Through a huge number of cameras mounted in public spaces, our public security organs"Skynet" monitoring system preserves a stable and safe social order. Many people also utilize home security cameras and pet monitors to ensure the safety of their homes and family members. 5) CONCLUSION: In this article, I've covered the basics of the edge computing model, including its architecture, important technologies, security, and privacy protection. Edge computing provides data storage and computation at the network's edge, as well as Internet intelligent services close by, assisting in the digital transformation of various businesses and satisfying the data diversification needs of various industries. Edge computing has become a popular topic of study. Edge computing will become more essential in the future as the Internet and human society continue to develop, effectively promoting the development of numerous sectors. It has a wide range of applications, including Content Delivery Networks (CDNs), industrial Internet, energy, smart homes, smart transportation, games, and more. 6) REFERENCES: 1. https://www.researchgate.net/publication/30612290 2. https://www.emc.com/leadership/digitaluniverse/2014iview/index.html 3. M. Satyanarayanan, ‘‘The emergence of edge computing,’’ Computer, vol. 50, no. 1, pp. 30–39, Jan. 2017 4. W. Shi, J. Cao, Q. Zhang, Y. Li, and L. Xu, ‘‘Edge computing: Vision and challenges,’’ IEEE Internet Things J., vol. 3, no. 5, pp. 637–646, Oct. 2016. 5. Z. M. Zha, F. Liu, and Z. P. Cai, ‘‘Edge computing: Platforms; Applications and challenges,’’ 6. W. S. Shi, X. Z. Zhang, and Y. F. Wang, ‘‘Edge computing: State-of-the-art and future directions,’’ J. Comput. Res. Develop., vol. 56, no. 1, pp. 1–21, 2019.