Abstract
Some of the largest global industries that is driving smart city environments are anywhere and anytime health monitoring applications. Smart healthcare systems need to be more preventive and responsive as they deal with sensitive data. Even though cloud computing provides solutions to the smart healthcare applications, the major challenge imposed on cloud computing is how could the centralized traditional cloud computing handle voluminous data. The existing models may encounter problems related to network resource utilization, overheads in network response time, and communication latency. As a solution to these problems, edge-oriented computing has emerged as a new computing paradigm through localized computing. Edge computing expands the compute, storage, and networking capabilities to the edge of the network which will respond to the above-mentioned issues. Based on cloud computing and edge computing, in this chapter an opportunistic edge computing architecture is introduced for smart provisioning of healthcare data.
Top1. Introduction
The evolution of the Internet of Things and technology has led to the unbelievable growth in the deployment of smart sensors, actuators, and low power consuming hardware chips, smart devices in various fields like telecommunication, manufacturing, aerospace, smart homes, smart city etc .Smart health care systems are one of the important fields that are witnessing this change (Sodhro, Pirbhulal, & Sangaiah, 2018). This development of smart environment creates a great burden on the network due to the enormous data transmission. This creates a challenge for the existing cloud infrastructure to provide timely service to the end users(Zhang et al., 2015). The burden that is put on the data processing and analytics on the cloud computing paved the way for the development of new computing paradigm that brings the compute, storage, and processing to the edge of the network that are closer to the user premises. The method of computing at the edge of the network is called “Edge Computing” (Yu et al., 2017).
According to the predictions that are made by (Koop et al., 2008), the present hospital-based health care systems will take a drift to hospital and home balanced by the year of 2020 and will eventually lead to home-based by the year 2030. To make this happen new architectures, technologies and new computing paradigms should be developed specifically to health care domain. Sending the data for computation to the cloud involves latency delay and health care applications are not tolerant of this delay. Hence Edge computing will provide solutions to this data intensive health monitoring system by reducing the network communication for data transfer, storage issues and latency. This chapter demonstrates the use of Edge computing wherein, the real-time data can be monitored, stored and later can be sent to other storages or clouds if required
Contribution of Edge Computing To Data Science and Analytics
Data science and analytics uses various methods, algorithms, and machine learning models to gain knowledge about the data that are analysed. Edge Computing has evolved to overcome many challenges and issues of cloud computing. They provide a way to make analysis and computation at the IOT domain level and at a level that are one step next to the IOT plane. Edge computing enables different stake holders and systems to perform analytics near to the users with the available resources. Developing various analytics and machine learning models at the edge level may reduce the computation time, response time, latency, bandwidth consumption and improve the quality of service.
Key Terms in this Chapter
Edge Computing: It is a distributed computing paradigm which brings computation and data storage closer to the needed location and near to the users, in order to improve response times, bandwidth consumptions.
Communication Latency: Latency refers to delays that is incurred during transmitting or processing data, which can be caused by varied reasons.
Machine Learning: Machine Learning models are scientific and statistical model that are developed on gathered data for classification, analysis, and forecast the future events.
Broker: The intermediate layer that are responsible for managing the resource pools.
Service Providers: A service provider is a vendor that provides IT solutions and/or services to end users and organizations.
Opportunistic Computing: Opportunistic computing use the resources, services, applications, and computing resources, contributed by the devices connected in an opportunistic network, for performing the execution of distributed computing tasks.
Bandwidth: Bandwidth describes the maximum data transfer rate of a network or Internet connection.