In an efficient way, clients using Web applications and batch processing can avoid owning and managing private data centers (DCs) by adopting the “pay-as-you-go” model of cloud computing. Fog Computing is a highly virtualized platform that sits at the edge of a network and offers networking, storage, and computation services between endpoints and standard cloud computing data centers. Storage, networking, and computation are the foundational elements of both the Cloud and the Fog. But the term “Edge of the Network” suggests several qualities that make the Fog a significant extension of the Cloud. A decentralized computing system called fog computing expands the capabilities of cloud computing to better support the Internet of Things (IoT) ecosystem. In order to give the networked smart devices that comprise the Internet of Things with faster and more localized computing services, it seeks to move intelligence, processing, and storage closer to the edge of the network. Although the phrases “edge computing” and “fog computing” are sometimes used synonymously, they differ slightly. Edge computing is the process of moving computing duties to the edge of the network, whereas fog computing acts as a computer layer between the cloud and the edge. Fog nodes receive data streams from the edge nodes and process them first. Based on their analysis, they determine whether to transfer the data to cloud nodes or carry out more processing on their own. Fog computing can be viewed as an extension of edge computing due to these factors. A highly virtualized Internet of Things node that facilitates networking, computation, and storage between edge devices and the cloud is called a fog device, sometimes known as a fog node. It shares traits with clouds in particular. It can be viewed as a mini-cloud that combines the enormous processing power of the cloud with its own resources and the data gathered from edge devices. Fog computing allows low-latency applications to be developed and deployed directly on devices like as routers, switches, tiny data centers, and access points by putting processing resources closer to end devices.

A new class of services called fog as a service (FaaS) is made available by fog computing and the Internet of Things. In this case, service providers establish geographically distributed arrays of fog nodes that function as rentable, ownerless resources for companies in a variety of industries. With its local networking, storage, and computing capacities, each fog node is a perfect choice for dispersed data processing and analysis. FaaS gives companies a new way to provide services to customers. FaaS offers both large and small enterprises public and private computing, control, and storage services, in contrast to the huge companies that usually operate clouds at different scales. The kind of directed data from the fog must be determined before the fog’s operation can be explained. At the network’s edge, developers can build fog node ports or Internet of Things apps. The data generated from IoT devices is absorbed by the fog nodes close to the network edge, and the fog IoT application subsequently routes this data to the best location for analysis.

The different types of data are as follows:

1. The data that is most time-sensitive is analyzed by the fog node in close proximity to the sources of the data, such as sensor systems.
2. Information that can wait minutes or even seconds for a reaction or action; this type of information is sent to an aggregation node for review, assessment, and subsequent action.
3. Less time-sensitive data are sent to the cloud for Big Data analytics, long-term storage, and historical analysis.

Application of Fog Computing

1. It is used in situations where sending a tiny amount of data to the cloud is necessary. This material is selected for long-term storage and is not commonly accessed by the host.
2. It’s used when a lot of services need to be provided at different locations and over a big area.
3. Devices that must withstand rigorous computations and processing must use fog computing.
4. IoT devices (like autonomous car, V2X), devices with sensors and cameras (IIoT, or the Industrial Internet of Things), and other applications use fog computing.

Fog computing removes the need for a round trip to the cloud for analysis, which speeds up awareness of and reaction to events. In the end, firms who use fog computing obtain quicker and more in-depth insights, which boosts business agility and improves safety.

—— Pinakshi De

Assistant Professor, DAITM