Edge-Computing Paradigm: Survey and Analysis on security Threads

The commencement of extensive applications of Internet of Things devices in information technology is generating massive amounts of data. The deployment of various Internet of Things devices/sensors within the complex interconnected networks gives rise to raw data from sensors, processed and controlled data, decision-making data providing intelligent solutions, etc. Internet of Things offers a common platform (called IoT cloud) for all the networks and devices connected to those networks. The analytics can be performed on data, and valuable information can be extracted. Massive data traffic generated by Internet of Things sensors and related processing poses an overwhelming load and cost on Internet of Things cloud related to bandwidth, latency and resource scarcity. This, in turn, degrades the quality of service and network performance. To cope with such issues, Edge Computing paradigms came into existence, extending the cloud storage capacity and computational resources close to specific Internet of Things devices. However, EC assisted Internet of Things to reduce the volume of data transition over the cloud but continued with significant risks associated with security and privacy. Moreover, the expansion of service requirements triggers security and efficiency issues.

I. Introduction

The beginning of the edge computing paradigm takes us back to the content delivery network (CDN) to seek the web-performance benefits. CDN can also be called a Content distribution network as it is the physically-distributed network of proxy servers, reverse proxy and their data centres. It uses edge nodes to serve the user request by prefetching and storing the cached version of web content using persistent caching in multiple geographical locations (CDN's Point of Presence). CDN opens the doors for emerging applications as it helps minimise the physical distance between the server and user and prevent site crashes during network overloading since it distributes bandwidth among multiple servers instead of allowing a single server to manage all the traffic.