Edge Computing

With the evident Cloud technologies’ domination of computing market, new demand conditions are materializing. A game changer in the last few years is the increase in the personal use of Cloud services, so much so that the next era is being dubbed the era of the personal Cloud. Advantages for clients are clear, and hence, the surge in client adoption rates is imperative. The personal Cloud combines all the connected devices (things), controlled by or servicing a single user or a house hold, to form a conceptual border around these elements within the Internet of Things (IoT) universe. This primarily affects Cloud/Internet of Things service providers as it is their Cloud resources that all of this load eventually flows into. Providers need enhanced vision of what needs to change in terms of how they handle Cloud management in general.

In this chapter, we strive to answer this question by investigating the areas which will be most affected in the Cloud management portfolio. The purpose is to offer a solid perspective into what challenges are expected in the environment, technical requirements and usage trends. This, in turn, would aid in constructing a vision for the directions providers need to go in with their policies and algorithm design.

Impacts on the area of resource allocation in the Cloud are discussed. This covers multiple venues including the demand and its impact on financial investment on Cloud resources, the optimality of the pricing plans, and the energy efficiency considerations in light of the personal Cloud conditions. Moreover, we offer a detailed discussion on Cloud client demand pattern prediction methods. A discussion is offered on the impacts of the intersection with Internet of Things performability. This delves into inherent IoT issues in the era of personal Cloud including challenges stemming from request volume, interconnectivity, privacy considerations, and legal issues. Recommendations on the Cloud provider strategy to tackle the service brokering challenge in a way that maximizes profitability from the stream of request coming from heterogeneous things are introduced toward the end.

Modern-day real-time IoT devices used in domains like automated surveillance, healthcare, augmented/virtual reality, automation and control etc are generating a huge amount of data and are very delay sensitive as well. Due to this, they are becoming bandwidth hungry and require an uninterrupted connectivity/communication channel as well. This gave birth to the use of small cells (micro, pico, femto) on the edge of the network to accommodate a large number of IoT devices. On the other hand, delay sensitivity of real-time IoT applications are forcing the adoption of Edge Computing rather than using a far Cloud. Edge Computing does process the sensed data near to its origin to meet the strict delay requirements. This chapter addresses these two issues and is trying to optimize Edge Computing and Edge Communication network using Integer Linear Programming (ILP). The ILP problem is formulated for optimal computation and communications are original and novel. Using ILP, an optimal way to utilize Edge Computing resources is proposed to meet the demand optimally. Similarly, it solves the issue of optimal and dynamic channel allocation (DCA) in small cells. DCA problem is also formulated as a novel ILP problem and solved.

With the exponential growth of smartphones, the growth of mobile traffic has also increased dramatically. With this, there has been also increase in the data involved – which is big data. A large part of big data is most valuable when it is analyzed quickly as it is generated. There is a need for processing continuous data streams under very short delays. Recently, frameworks and architectures have been proposed for carrying out data stream processing at the edge of the network using constrained resources. This chapter aims to present a comprehensive survey of the framework, architecture, and applications areas in the area of mobile edge computing. It also discusses some of the challenges and related existing solutions as well. It also provides a survey of the state-of-the-art mobile edge computing research with the focus on deep learning as a technique used for reliable and secure deployment of MEC.

The Internet of Things (IoT) is expected to grow faster than any other category of connected devices. IoT allows any device with an on-and-off switch to connect to the internet—a concept that has the ability to greatly change our lives and work. These modern systems collect inherently complex data streams due to the volume, velocity, value, variety, variability, and veracity of data, which leads to incremental growth in data traffic on networks and in the cloud. To fulfill the requirements of IoT, including geodistribution, low latency, location awareness, and mobility support, a new paradigm is proposed: edge computing. In edge computing, substantial computing and storage resources are placed at the edge of the network in mobile devices or sensors. The term “edge” is taken from network diagrams; normally, the edge of a network diagram represents the point at which data traffic enters or leaves the workable network. Using the concept of edge computing, an organization can shift huge amounts of data into processed data near the data origin, which helps to reduce data traffic in the network’s central repository (called the “cloud”). Edge computing uses a variety of data reduction techniques close to the data source at the network edge, including data pre-processing, local storage, and filtering. This approach can prevent some critical issues, such as I/O bottlenecks, storage and bandwidth limitations, data traffic increments, and high energy costs. A major advantage of edge computing is improvement of the request-response delay to milliseconds. Edge computing also supports security and network challenges. However, two major obstacles exist toward achieving the benefit of network-edge computing. First, the most efficient algorithms for data reduction in time series (one of the most common types of data in IoT) were developed to work posteriori upon big datasets, but they cannot make decisions for each incoming data item. Secondly, the state of the art lacks systems that can apply any of the possible data reduction methods without adding significant delays or major reconfigurations. Edge computing has also inherited some of the challenges of cloud computing, including data abstraction, naming, and programmability. This chapter presents a detailed taxonomic discussion of edge computing, along with its challenges, opportunities, and data reduction methods.

In the modern world, the Internet of Things (IoT) has captured most of the applications in the market and it plays a significant role with various smart and interactive things that have provided a convenient environment for humans. Applications of IoT can be broadly classified based on consumer application, enterprise application, infrastructure management, industrial productivity, and so on. Some of the applications that implement the concepts of IoT are smart wearables, connected cars, industrial internet, precision agriculture, smart retail, energy management systems, and others. Advancement in IoT technology has resulted in automated home appliances and healthcare systems and also in the development of smart cities, smart homes, smart towns, and many more. For any application based on IoT, data are collected and are modeled using prediction analysis. The basic concept of using IoT is that it reduces waste, loss, and cost. The design of the IoT application should focus on increasing the energy efficiency and the cost efficiency of the system, improving safety, or creating better experiences. IoT is a field of the network, and it could be enabled either by wired or wireless communication networks. This refers to the technology where the machines could predict the data and work by itself. Here the devices that are connected to the private internet connection could communicate with others. The edge devices that are connected will create a huge amount of data. These data are generated at high speed, and also they face challenges like storage, computation, and networking. These issues can be handled by using cloud, edge, or cloud edge computing. This chapter focuses on the discussion of diverse applications in IoT because of its huge potential market, and also it enhances the comfort of our lives besides providing enhanced control of our daily and personal tasks with ease.

Software-Defined network (SDN) attracted plenty of researchers from various technological fields who have contributed to enhance the network. SDN is a highly advanced technology which makes it easy for engineers to update protocols and other parameters at runtime (without switching off the devices). Recently, smart cities concept has been introduced, where devices in multidirectional form will be connected to provide timely and useful information to all kind of people and government. A number of researchers have attempted to merge SDN and IoT to provide better information to users. In this chapter, a novel concept has been introduced to combine both these technologies through a software-defined things architecture. There are many advantages of the proposed architecture where all data services are further connected via two intermediate levels working on SDN principles. Both the abovementioned technologies have a great potential for smart cities projects. The proposed architecture is evaluated using Spark and GraphX with Hadoop ecosystem which showed encouraging results especially the efficiency of real-time transfer of data over SDN.

The advancement and convergence of Internet of Things (IoT), mobile devices technology, big data and cloud computing with its various technological implementations are finally enabling the vision of Smart Cities and Industry 4.0. However, cloud computing concept has been built with the assumptions of good network connectivity, adequate bandwidth and low latency. But with the proliferation of interconnected smart devices and the expected huge amount of traffic and data to be generated, coupled with the stringent and extremely demanding connectivity, high bandwidth and low latency requirements placed on applications and services were embedded in Smart Cities and Industry 4.0 concepts. The traditional cloud-centric architectural arrangement no longer holds due to these cloud architectural model assumptions. Cloud computing is therefore gradually evolving into new complementary concepts as edge and fog computing, and now mist and dew computing. Mist computing addresses these concerns by extending the capabilities and features of cloud and fog computing, with some level of computing intelligence further on the extreme edge of the network closer or on the sensing devices. With the technological revolution currently spreading across Africa, policy makers, academics and businesses in Africa are gradually recognizing the potential opportunities embedded in embracing emerging and future technologies to tackle issues related to urbanization and industrialization as catalyst for sustainable development and growth. This chapter studies the current trend in mist computing and discusses the application and the potential use case scenarios for Smart Cities and Industry 4.0 in the context of Africa. The chapter also explores practical implementation challenges and drivers supporting growth of these emerging cloud technologies in the region. Finally, pertinent technical recommendations were proposed as solution to the challenges identified together with a qualitative analysis of future opportunities of mist computing in the overall vision of Smart African Cities and Industry Africa 4.0.

One of the applications of big data research is to utilize inexpensive and unobtrusive Internet of Things- (IoT) driven devices for monitoring hospitalized patients whose physiological status requires close attention. This type of solution employs sensors to collect physiological information and uses gateways to send the data or warnings to caregivers for further analysis. Unfortunately, real-world applications of health monitoring for mobile users were so far poor mainly due to the energy constraints imposed by the batteries. Edge computing aims to process data produced by devices to be closer to its origin instead of sending it to data centers.

This chapter presents a ZigBee-based gastrointestinal track motility monitor (GTMM), an IoT-driven eHealth device that is specifically designed for constant monitoring of hospitalized patients after major abdominal surgery. GTMM after abdominal surgery is required for preventing unexpected postoperational complications such as intestinal obstruction.

Cloud computing eliminates the advantage, which large organizations have conventionally enjoyed in terms of the availability of technology specialists and technical superiority. Small and medium enterprises that take advantage of infrastructure providers to support their technology requirements and provide specialized platforms for development and testing can build an infrastructure that is innovative and is capable of enabling entry into a global marketplace with as much capability as the market demands. Smaller enterprises have the ability to leverage software services that provide software such as supply chain management, enterprise resource planning, customer relationship management, and business analytics which are traditionally available only to large enterprises and organizations. Being able to access infrastructures, platforms, and software services based on what is needed and paying for only what is used enable and empower start-up enterprises and small and medium enterprises, giving them an advantage in the market and also an equal position with much larger enterprises.

IoT can be viewed as networks of networks. There can be a wide range of applications in IoT that supports logistics and supply chain management. IoT technology can be leveraged to achieve cost reductions. IoT technology can be combined with real-time location systems to get live updates from the factory floor, enabling manufacturers to continuously monitor machine activity, maintenance needs, and also product movement during production. Cost reduction can be achieved across the digital supply chain by making use of these smart machines by providing data that allow manufacturers to adjust production on the fly. Manufacturing and assembly lines will receive updated schedules and quality-related information in real time and instantly. IoT data can be leveraged to schedule proactive, preventive, predictive repairs, maintenance, customize production to meet the customer’s orders, and the focus that is needed to be successful in the digital world. The concept of Industry 4.0 aims at achieving smart factory will soon be a reality. Smart products which consist of the embedded knowledge of their customers’ needs will provide data insights and analytics about the best way to achieve customer fulfillment. All this information will lead to more cost-efficient production and product development.

Digitally enabled real-time collaboration partners will need to collaborate across all nodes of the supply chain to profitably meet the customers’ demands. Select technology solutions so that the supply chain partners can work within and across various networks and at touch points. Supply chain management (SCM) manages to optimize processes and collaboration with other companies in the supply chain (suppliers and customers) to create more value. While SCM is already heavily supported by various IT solutions, the Internet of Things (IoT) can be of great value by providing additional information. One of the major challenges in SCM is reducing the bullwhip effect. A major cause of the bullwhip effect is information distortion. For a better information flow, the IoT is able to trigger all relevant actors in the supply chain upon the sale of a product. In traditional processes, information on demand was only passed to one’s direct downstream partner instead of sharing this information with the whole chain. IoT can enable sharing of information across the entire supply chain from the upstream suppliers to the downstream customers.

In the previous decade, Internet of Things (IoT) has paved ways for connecting devices, people, infrastructure, and practically any entity that can be connected. Cloud infrastructure has been the irrefutable part of IoT solutions with substantiated claims of resource provisioning, scalable platforms, and minimal infrastructure costs. Cloud computing does not argue well with applications that require reduced latency, positioning capabilities, and support for mobility. In such instances, edge computing can be used to compliment the cloud infrastructure and help in mitigating the various technical bottlenecks that are part of cloud services. In IoT environment, edge computing can help in improving energy efficiency, optimal resource utilization, and contextual data processing and in reducing the network traffic of backbone networks. The characteristics of edge computing may lead to new challenges for ensuring security and privacy. This chapter proposes a conceptual framework which consists of the essential components to ensure security and privacy for edge computing applications.