An Overview on Edge Computing 1

1. INTRODUCTION

1.1 Objectives
1. To look into the edge architecture and understand the transmission of data entities from
the various nodes.
2. To understand what edge computing is and its real time applications in the real world.
3. To compare edge computing and cloud computing algorithms to gain an insight into
their respective pros and cons.
4. To understand the future scope and direction of edge computing.

1.2 Problem Statement

To get acquainted with a new computer paradigm known as ‘Edge Computing’ in
order to understand its utility and how it functions as an extension of cloud computing.

Department of Computer Engineering

 An Overview on Edge Computing 2

2. LITERATURE SURVEY

1. An Overview on Edge Computing Research

Year – 2020;
Authors - Keyan Cao, Yefan Liu, Gongjie Meng, Qimeng Sun

With the rapid development of the Internet of Everything (IoE), the number of smart
devices connected to the Internet is increasing, resulting in large-scale data, which
has caused problems such as bandwidth load, slow response speed, poor security,
and poor privacy in traditional cloud computing models. Traditional cloud
computing is no longer sufficient to support the diverse needs of today's intelligent
society for data processing, so edge computing technologies have emerged. It is a
new computing paradigm for performing calculations at the edge of the network.
Unlike cloud computing, it emphasizes closer to the user and closer to the source of
the data. At the edge of the network, it is lightweight for local, small-scale data
storage and processing. This article mainly reviews the related research and results
of edge computing. First, it summarizes the concept of edge computing and
compares it with cloud computing. Then summarize the architecture of edge
computing, keyword technology, security and privacy protection, and finally
summarize the applications of edge computing.

2. ScalEdge: A framework for scalable edge computing in Internet of

things–based smart systems
Year – 2021
Authors – Mohammad Babar, Muhammad Sohail Khan

Edge computing brings down storage, computation, and communication services

from the cloud server to the network edge, resulting in low latency and high
availability. The Internet of things (IoT) devices are resource-constrained, unable

Department of Computer Engineering

 An Overview on Edge Computing 3

to process compute-intensive tasks. The convergence of edge computing and IoT

with computation offloading offers a feasible solution in terms of performance.

3. The Edge Completes the Cloud Methodology

Year – 2020
Authors - Bob Gill, David Smith

Edge computing augments and expands the possibilities of today’s primarily

centralized, hyperscale cloud model, supports the systemic evolution and deployment
of the IoT, and supports enabling next-generation digital business applications.

4. Edge Machine Learning for AI-Enabled IoT Devices: A Review

Year – 2020
Authors - Massimo Merenda, Carlo Porcaro, Demetrio Iero

In a few years, the world will be populated by billions of connected devices that will be
placed in our homes, cities, vehicles, and industries. Devices with limited resources will
interact with the surrounding environment and users. Many of these devices will be
based on machine learning models to decode meaning and behavior behind sensors’
data, to implement accurate predictions and make decisions. In this work, a detailed
review on models, architecture, and requirements on solutions that implement edge
machine learning on Internet of Things devices is presented, with the main goal to
define the state of the art and envisioning development requirements.

5. A Smart Manufacturing Service System Based on Edge Computing,

…..Fog Computing, and Cloud Computing
Year – 2020
Authors – Q. Qi, F. Tao,

Department of Computer Engineering

 An Overview on Edge Computing 4

In a smart manufacturing environment, more and more devices would be connected

to the Internet so that a large volume of data can be obtained during all phases of
the product lifecycle. Cloud-based smart manufacturing paradigm facilitates a new
variety of applications and services to analyze a large volume of data and enable
large-scale manufacturing collaboration.

6. Priority-based residential energy management with collaborative

edge and cloud computing
Year – 2019
Authors - L. Ruan, Y. Yan, S. Guo, F. Wen, and X. Qiu

To improve latency and processing performance, a three-tier edge-cloud

collaborative REM (ECCREM) architecture is presented. In consideration of
matching the architecture, a two-stage energy management mechanism is proposed
with system reliability and resource utilization requirements taken into account. In
addition, energy cost can be reduced and demand fluctuation relieved.

7. Comparison of edge computing implementations: Fog computing,

cloudlet and mobile edge computing
Year – 2017
Authors – Koustabh Dolui, Soumya Kanti Datta

With the massive growth in intelligent and mobile devices coupled with
technologies like Internet of Things (IoT), V2X Communications, Augmented
Reality (AR), the focus has shifted towards gaining real-time responses along with
support for context-awareness and mobility.

Department of Computer Engineering

 An Overview on Edge Computing 5

3. Architecture of the System……….

3.1 Architectural Classification in Edge

The three factors, namely, data flow, control, and tenancy, serve as the basis of
classifying the architectures for management of resources in edge computing.
1) Data flow architectures - It is directed towards the workload’s movement and data
inside the processing ecosystem. For instance, workloads are portable to the edge nodes
from the user devices or cloud servers to the edge nodes.
2) Control architectures: It is dependent on the way of controlling resources in the
processing ecosystem. For instance, various edge nodes are managed by a principal
algorithm or a central controller. The distributed approach is another option that can be
used.
3) Tenancy architecture: It relies on the reinforcement given for providing numerous
entities in the computing ecosystem. An edge node can host a solo application or a
group of applications.

3.2 Description of the Peripheral Architecture

Edge computing architecture provides us a better replacement to cloud computing
architecture. Edge computing circumscribes an ecosystem of architectural components
that have been distributed from the central location of the data centre to across all the
nodes reachable.

Fig. 1 – Nodes of edge computing

Department of Computer Engineering

 An Overview on Edge Computing 6

Fig. 1 gives us the broad spectrum of edge device connectivity. This is the external
architecture of edge computing. Presented here are the four scales of data
containment, Enterprise Hybrid Cloud, Edge Cloud, Edge Server/Gateway and
Edge Devices. The smaller facets form the subsets of the larger components,
ultimately leading to interlinking of data, whilst maintaining absolute privacy.

1) Enterprise Hybrid Cloud

An enterprise hybrid cloud is an amalgamated setting that combines the abilities of
public and private clouds. It also integrates security and control of data centers.
Additionally, this model provides a unified and centralized means of management
and governance across all systems—cloud and on-premise. It contains the following
keys:
i) Enterprise user directory – user directory for enterprise cloud which can be
replicated to edge cloud.
ii) Enterprise data - data aggregated from sources and stored within the
enterprise for future analysis, audit, and history.
iii) Device registry - stores information about devices that the edge system can
read, communicate with, control, provision, or manage.
iv) Model repository - stores Al models for reference and further analysis.
v) Model Management Service - Model Management Service that supports
storage, delivery, and security of models and other metadata packages.

2) Edge Cloud
An edge cloud architecture is used to reorganize (processing) power to the edges
(clients/devices) of your networks. Conventionally, the computing power of servers
is used to perform tasks such as data minimization or to create cutting-edge
distributed systems. Within the cloud model, ‘intelligent’ tasks are completed by
servers so they can be relocated to and from other devices and the enterprise hybrid
cloud edge with less or almost no computing power.

3) Edge Server/Gateway
An edge gateway accesses and makes required changes to the data from edge
devices, sending back appropriate data, providing network translation between
networks using different protocols, thus resulting in a drop in the bandwidth used.
The edge gateways help reduce technology compromise and help in boosting
security of the network.

4) Edge Devices
An edge device can be considered to be any components of computer hardware that
control data flow at the edge gateway. Edge devices satisfy a variety of tasks,
depending on what type of device they are, but they fundamentally serve as network
entry or exit points.

Department of Computer Engineering

 An Overview on Edge Computing 7

3.3 Flow of the Data

The general flow diagram is given below -
1) The Edge Application Manager for Devices is installed on a Kubernetes cluster.
Devices are registered and authenticated directly within Horizon exchange.
2) Agents are installed on all edge devices that support ARM and other architectures
that run Docker. These agents communicate with other components to securely
orchestrate software lifecycle management on their machines.
3) In the reference architecture, the focus is on the outermost edge, the edge devices
region. Edge workloads are microservices that are packaged as containers, either
individually or as a collection.
4) Workloads are placed on individual nodes as the result of an agreement that is
negotiated between the device agent and an agreement-bot.
5) The agent and the agreement-bot actively monitor that the negotiated agreements
remain valid in the presence of changing conditions.
6) The Model Management Service is associated with agents where it coordinates the
synchronization with the devices. Another instance of Model Management Service
is in the enterprise region, which is the central management hub that masters the
original models.

Fig. 2 – Representation of External Architecture in Node Form

Department of Computer Engineering

 An Overview on Edge Computing 8

3.4 Description of the Central Architecture

The model-driven united service framework is located at the top of the scale to
realize the development and deployment of services. According to the general
framework of edge calculation, it is divided into cloud, edge layer and field layer.
The edge layer consists of two main parts: the edge node and the edge manager.

Fig. 3 – Internal Description of an Edge Node

1) Cloud Layer

A number of high-performance servers and storage devices are present in the cloud
computing layer, with powerful computing and storage potential. This layer is
important to process requests involving large amounts of data analysis such as
regular maintenance and business decision support. The cloud computing epicenter
can permanently store the reported data of the edge computing layer. Furthermore,
the cloud module can flexibly alter the deployment strategy of the edge computing
layer.

Department of Computer Engineering

 An Overview on Edge Computing 9

2) Edge Layer
The central edge layer is the fundamental section of this architecture. It is positioned at
the edge of the network, consisting of edge nodes widely distributed between terminal
devices and clouds. The edge layer validates the contact of terminal devices downward,
and simultaneously stores and computes the data uploaded by nodal devices. Usually
contains base stations, access points, routers, switches, gateways, etc.

3) Terminal/Entry Layer
The terminal layer is the outermost layer and consists of all types of devices connected
to the edge network, including mobile workstations and many Internet of Things
devices (such as sensors, smartphones, smart cars, cameras, etc.). In order to reduce the
terminal service delay, only the perception of the various terminal devices is
considered, not the computing power. As a result, hundreds of millions of devices in
the terminal layer collect all kinds of raw data and upload it to the upper layer, where it
is stored and calculated. The terminal is not only a data consumer but a data provider.

3.5 Design Goals of Architecture

The frame of reference is based on model-driven engineering method. In order to model
the knowledge of the physical and digital world, we need to achieve the following four
goals:
1) Launch a real-time and systematic cognitive model of the physical world and
achieve the cooperation between the physical world and the digital world.
2) Establish recyclable information model system in each vertical industry based on
modelling method.
3) System to system, service to service, and other model-based interface for
interaction, to achieve decoupling of software interface and development language.
4) Can effectively support the life cycle of development service, deployment
operation, data processing and security.

Department of Computer Engineering

 An Overview on Edge Computing 10

4. Technology Used in the System

4.1 What Edge Computing Enables

Edge computing is not a technology looking for a market, but a topology and design
being fulfilled with technologies such as IoT and the cloud to enable new solutions.
In particular, edge topology supports the placement of computing along a
continuum that ranges from a centralized isolated data centre at one extreme, to an
edge device equipped with machine learning and analytics capabilities closer to the
edge. By moving data and processing power closer to creators and users of content,
whether people, applications or things sensing their environments, we maximize or
balance the degree to which each of the edge’s attributes is satisfied.
We use five different categories of benefits, or imperatives of the edge, to determine
how beneficial edge computing may be for a given application. We group these
imperatives as:
1. Data volume/bandwidth considerations
2. A need for limited autonomy or disconnected operation
3. Privacy and security concerns
4. A requirement for local interactivity
5. The effect that latency may have on an application

4.2 Algorithms Formulated

The four fundamental algorithms enabling edge computing are:
1) Discovery

Discovery algorithm is used to get the ideal or the most optimized model among
several thousands of pre-designed models available. Several executions of
different discovery algorithms could be required trying to obtain a quality
model. Several techniques execute different discovery algorithms for an event
log and evaluate their resulting models using quality metrics. Here, it spends its
time finding edge resources inside the network to be utilized for distributed
processing.

2) Benchmarking

Benchmarking algorithms run an algorithm on a set of test functions and extract

performance measures from the generated data, thus acquiring resources’

Department of Computer Engineering

 An Overview on Edge Computing 11

performance to make decisions in expanding the performance concerning

implementation of edge networking.

3) Load-balancing
Load Balancer is the component in distributed systems that spreads the incoming
requests from the client to cluster of servers. This ensures that if one server goes
own, the load balancer will redirect that request to any other server. So, it can
maximize the responsiveness and availability of the application. In edge computing,
load balancing is used in distributing workloads among various resources with some
bases like fairness, priorities, etc.

4) Placement
Placement algorithms are resource finders and allocators in any distributed system.
In edge computing implementation, all it does is finding resources necessary for the
deployment of workload.

Table 1 presents the techniques used and applications based on the four algorithms.

Table 1 – Types of Algorithms

Algorithms Techniques used Services Application areas

Discovery Programming Foglets, Public resource
frameworks, Edge-as-a-Service (EaaS), distribution.
protocols for Internet. Online game.
handshaking and simulation-based
message passing validation
Benchmarking Basic conventions for Spark, IoT application
assessing functional Cloud Sim
properties,
benchmarking based
on application and
integrated
benchmarking
Load-balancing Particle Swarm Internet of vehicles, Effective load
Optimization - Cooperative load-balancing balancing
Constrained model authenticated data
Optimization Breadth-First Search (BFS) centre.
Placement Dynamic condition- Priced timed Petri-nets Dynamic behaviour on
aware (PTPNs) performance, time and
iterative techniques Fog-to-Cloud (F2C) cost.
Real-time applications

Department of Computer Engineering

 An Overview on Edge Computing 12

4.3 Motive for Algorithm Formulation

1) Proximity is in the edge: This is the old but still valid argument of peer-to-peer
(P2P) systems and content distribution networks (CDNs). It is more efficient to
communicate and distribute information between close-by nodes than to use far-
away centralized intermediaries.
2) Intelligence is in the edge: As miniaturization still continues and computing
capacity still increases, edge sensors and devices become more powerful. This
opens the way to autonomous decision-making in the edge such as novel distributed
crowdsensing applications, but also human-controlled actuators or agents reacting
to the incoming information.
3) Trust is in the edge: Personal and social sensitive data is clearly located in the edge.
The control of trust relation and the management of sensitive information in a secure
and private way must therefore also belong to the edges.
4) Control is in the edge The management of the application and the coordination also
comes from the edge machines that can assign or delegate computation,
synchronization or storage to other nodes or to the core selectively.
5) Humans are in the edge: Human-centered designs should put humans in the control
loop, so that users can retake control of their information. This should lead to the
design of novel crowdsourced and socially informed architectures where users
control the links of their networks. Finally, is also opens opportunities for novel and
innovative forms of human-centered applications.

4.4 Technical Execution

Fig. 4 shows, using UML 2.0, the elements and relations that determine the scope
of this theory. We describe the theory scope through four archetype classes: Actor,
Technology, Activity, and Software System. The four archetype classes have been
represented as abstract classifiers (classes). We added several classifiers
(subclasses) to indicate that the activities of IoT edge systems are performed on the
edge side.

Specifically, the subclass IoT_Edge_Computing represents a technology

understood as a set of skills, techniques, methods, and processes, all specialized for
the IoT edge computing paradigm. The subclass Activity_IoT_Edge represents an
activity performed at the edge; in addition, this class has been declared as active,
since, by its very nature, its instances will have their own control flows. Finally, the
subclass IoT_Edge_Software_System represents a software system in the IoT Edge
computing domain.

Department of Computer Engineering

 An Overview on Edge Computing 13

Fig. 4 – Relational Diagram

Fig. 5 shows the relations between the elements of the scope and the constructs with
respect to edge computing. Regarding this figure, class Device_in_the_Edge and its
subclasses represent the constructs C1 to C4. The classes Sensor and Actuator
represent the construct C5. Distributed_Architecture represents the construct C6. The
classes Advantage and Problem represent constructs C7 and C8, respectively.
Construct C9 (represented by the classifier Activity_IoT_Edge) is an artifice (it is not
central to the theory, although it is part of the scope of the theory) that allows us to
establish two levels of abstraction in the operations performed by an
IoT_Edge_Software_System. These high-level operations generate the benefits of the
IoT_Edge_Computing technology. For example, the storage and analytics in the
Device_in_the_Edge, and the filtering and artificial intelligence techniques enable
local data processing and avoid sending raw data to fog/cloud, gaining better
bandwidth throughput. Table 2 relates the constructs to the classifiers in Fig. 5.

Table 2 - Relations between constructs and classifiers

Construct Classifier
C1 - Device_in_the_Edge 1. Device_in_the_Edge
C2 - Device_in_the_Edge:: type
C3 - Device_in_the_Edge:: attributes
C4 - Device_in_the_Edge:: functionalities
C5 Physical elements 2. Sensor, Actuator
C6 Architecture 3. Distributed_Architecture
C7 Concerns::Benefit 4. Advantage
C8 Concerns::Challenge 5.Problem
C9 Activity 6. Activity_IoT_Edge

Department of Computer Engineering

 An Overview on Edge Computing 14

Fig. 5 – Constructs and scopes of theory

Department of Computer Engineering

 An Overview on Edge Computing 15

5. Analytical Study

5.1 Basic Understanding of Technology

After studying the technology of edge computing, we come across several reasons to
apply this technology in real life applications. However, for various reasons, the cloud
computing paradigm is unable to meet certain requirements (e.g. low latency and jitter,
context awareness, mobility support) that are crucial for several applications (e.g.
vehicular networks, augmented reality). To fulfil these requirements, various
paradigms, such as fog computing, mobile edge computing, and mobile cloud
computing, have emerged in recent years.
While these edge paradigms share several features, most of the existing research is
compartmentalized; no synergies have been explored. This is especially true in the field
of security, where most analyses focus only on one edge paradigm, while ignoring the
others.

5.2 Weighing the Problems and Solutions

1) Problem - Some issues created in edge computing like security (because there is no
….central device), and performance and battery issues in the mobile devices
Solution - A new Hybrid Mobile Edge Computing (HMEC) model is presented
Advantages -
1) The applied strategy is that, the performance cost model would be 70% and the
energy-saving cost model would be 30%, as long as the battery is above 50%.
And then when the battery is below 50%, the performance cost model starts to
decrease gradually to extend the battery lifetime.
2) There is also this improvement in HMEC that, the trust intermediary unit is
made between the mobile device and the processing unit utilized for offloading

2) Problem - Problems exist in application development in the domain

Solution - The Container based solution using remote debugging at the edge. This
container provides app developers, the ability to write codes in a productive
environment
Advantages -
1) Use case CNN (Convolutional Neural Network), whose purpose is to check the
performance of the computer’s vision-task.
2) sequential image layer downloading

Department of Computer Engineering

 An Overview on Edge Computing 16

3) Decompression
4) I/O Pipelining

3) Problem - The current system could not handle the IoT applications, its main
drawback being that it didn’t respond at the real-time.

Solution - An IoT device, which processes physical data and manages the devices
and gives real-time response
Advantages -
This device would provide the roadside assistant system; by using that system, the
consumers would be able to speak and their voice would be sent to the service
providers, and that way the consumers will get the quick response

4) Problem - Traffic issues increase in the browser

Solution - The two protocols coded MUTP, and Multi-source transmission protocol
are introduced for the downloading of videos on mobile devices
Advantages –
The Uncoded MUTP works double fast than the Parallel activity. And the coded
MUTP works three times faster than the HTTP and 25% faster than the Uncoded
MUTP for downloading the video; because these are using the Origo server.

5) Problem - A huge amount of data is processed in the IoT, and this data puts a lot of
burden on the cloud

Solution - Edge computing device is proposed

Advantages –
ECD device, the burden on the cloud is reduced, and also, it filters the data and then
sends it forward.

6) Problem - The edge devices are very limited. Due to the dynamic nature of the IoT,
too much uncertainty arises and due to that problem, this design is not suitable for
the latency-sensitive applications’ requirements and the privacy of user’s data

Solution - A decentralized resource management framework is proposed

Advantages –
This framework utilizes the available resources easily and efficiently; add to that, it
also guarantees the privacy protection.

Department of Computer Engineering

 An Overview on Edge Computing 17

5.3 Analytical Inference

1) Geographical Distribution
IoT applications based on sensor networks highly benefit from processing data
locally through edge computing platforms. Big data analytics can be performed
rapidly with better accuracy. Edge systems enable real-time analytics and AI
processing on a large scale. Examples include sensor networks to monitor the
environment, for example, pipeline monitoring or collision avoidance systems.

2) Mobility Support
As the number of mobile devices is rapidly growing, Edge computing also supports
mobility to communicate directly with mobile devices.

3) Location Awareness
Allows the use of technologies such as GPS to find the location of devices. Hence,
location awareness can be used by Edge computing applications such as Fog-based
vehicular safety applications and edge-based disaster management.

4) Proximity
The availability of the computational resources and services in the local vicinity
allows the users to leverage the network context information for making offloading
decisions and service usage decisions.

5) Low Latency
The low latency of Edge computing enables the users to execute their resource-
intensive and delay-sensitive applications on the edge device. Such applications
include connected vehicles, remote health monitoring, warehouse logistics, and
industrial control systems.

6) Heterogeneity
It refers to the existence of different platforms, architectures, computing, and
communication technologies used by the Edge computing elements (end devices,
edge servers, and networks). Bandwidth intensive use-cases. An increasing amount
of data generated by IoT deployments today is bandwidth-intensive, especially the
data from video from surveillance cameras (video analytics). Placing computational
resources as close as possible to high-bandwidth data sources implies that much less
data need to be sent to the distant cloud data centers. For instance, videos and sensor

Department of Computer Engineering

 An Overview on Edge Computing 18

data from hazardous locations can be processed locally to provide real-time

information to responders in public safety applications.

Department of Computer Engineering

 An Overview on Edge Computing 19

6. Experimental Implementation……

6.1 Data Collection

In IoT, massive amounts of data are collected at the edge of the network, but not all of
it is useful. If the data isn’t changing significantly, that means the servers are working
well. In the past, companies would send all of their monitoring data into the cloud or
to a corporate data centre for processing, analysis and storage. As the IoT has grown,
however, the volume of data makes this approach impractical. Edge compute performs
processing close to where the data originates. That can greatly reduce or even eliminate
the cost of the bandwidth needed to transmit it to the cloud or the corporate data center.
Some applications do need to examine data at the edge. An intelligent or AI-enabled
edge compute process can then immediately assess whether the situation demands a
response in real time, or send it on to the data centre for analysis.
Data collected at the edge falls into roughly three types:
1) It needs no further action and does not need to be stored.
2) It should be retained for later analysis and/or record keeping
3) It requires an immediate response

6.2 Key steps to implement edge computing

1) Decide how much intelligence there will be in your IoT devices.
As the data is filtered at the source, therefore, the more intelligence per device, the less
of it will actually be needed in the edge servers. Intelligent, standardized IoT devices
will provide lower volumes of data in more easily managed formats. However,
intelligent IoT devices have an associated higher cost; it's important to find a happy
medium.

2) Decide how you are going to group your IoT devices.

This might be decided to some extent by how you deal with step one -- a disparate
grouping of intelligent IoT devices might be easier to manage than one of a collection
of relatively dumb devices, as less filtering, analysis and reporting will have to be done
on the data streams.
Therefore, it might be preferable to group devices by proximity, rather than capability.
Proximity grouping also lowers latency and enables far faster response to identified
events. Again, this means that the edge server must be more intelligent, as it will deal
with different data streams reporting on different events.

3) Define carefully what the preferable outcomes are.

Department of Computer Engineering

 An Overview on Edge Computing 20

It's tempting to try to use edge systems to completely carve out areas of a platform
and to use these servers to fully manage the IoT devices, but you shouldn't. A
monitored high reading in one IoT device through one edge server might be
meaningless on its own, for example, yet when compared to similar devices being
monitored through all edge servers, it might be incredibly important. As such, it's
necessary to define what a true exception is and how such exceptions must be
handled.

4) Use a hub-and-spoke approach.

To manage the flow of data that's required, you must have an edge infrastructure
that consists of different edge servers placed across a network with a hierarchical
manner of dealing with the data between them. The optimum way to deal with such
a complex system is to have the lowest-cost, least-intelligent edge servers -- a
relative use of terminology, these systems might be quite intelligent and costly in
themselves -- as close to the IoT devices as possible.
An example here could be in a data centre where one edge server picks up a high
temperature reading. As far as it's concerned, it's a local event, but it sends that event
through to the central server. That server requests that all other outer edge servers
send through readings from all appropriate temperature monitors. If they are all
within limits, then yes, it's a local problem, probably due to a single item
overheating. However, if other reports come through of even minor increases in
temperature, it might mean that the cooling system for the whole data center has
failed -- requiring a much different set of events to correct.

5) Employ advanced data analytics and reporting.

Although automation has come a long way, it's still not 100% accurate. As such,
there will be cases -- there could be quite a lot of them due to the lack of maturity
of the area currently -- where edge servers must alert a human to an action to be
taken. False positives and negatives must be avoided and possible routes to
remediation must be shown to any human that gets involved. As such, don't scrimp
on the analytics tools used, and make sure that reporting is carried out in a clear and
meaningful manner.

Department of Computer Engineering

 An Overview on Edge Computing 21

7. Applications……………

Edge computing brings processing and storage capabilities closer to where it is needed.
Some of the more important applications of edge computing are presented below

1) Autonomous vehicles
Autonomous truck convoys will likely be one of the first use cases for autonomous
vehicles, involving a group of trucks travelling close behind one another in a convoy,
saving fuel costs and decreasing congestion. With edge computing, it will be possible
to remove the need for drivers in all trucks except the front one, because the trucks will
be able to communicate with each other with ultra-low latency.

2) Remote monitoring of assets in the oil and gas industry

Oil and gas failures can be disastrous. Their assets therefore need to be carefully
monitored. However, oil and gas plants are often in remote locations. Edge computing
enables real-time analytics with processing much closer to the asset, meaning there is
less reliance on good quality connectivity to a centralised cloud.

3) Smart grid
Edge computing will be a core technology in more widespread adoption of smart grids
and can help allow enterprises to better manage their energy consumption.
Sensors and IoT devices connected to an edge platform in factories, plants and offices
are being used to monitor energy use and analyse their consumption in real-time. With
real-time visibility, enterprises and energy companies can strike new deals, for example
where high-powered machinery is run during off-peak times for electricity demand.
This can increase the amount of green energy (like wind power) an enterprise consumes.

4) Predictive maintenance
Manufacturers want to be able to analyse and detect changes in their production lines
before a failure occurs. Edge computing helps by bringing the processing and storage
of data closer to the equipment. This enables IoT sensors to monitor machine health
with low latencies and perform analytics in real-time.

Department of Computer Engineering

 An Overview on Edge Computing 22

5) In-hospital patient monitoring

Healthcare contains several edge opportunities. Currently, monitoring devices (e.g.

glucose monitors, health tools and other sensors) are either not connected, or where
they are, large amounts of unprocessed data from devices would need to be stored
on a 3rd party cloud. This presents security concerns for healthcare providers. An
edge on the hospital site could process data locally to maintain data privacy. Edge
also enables right-time notifications to practitioners of unusual patient trends or
behaviours (through analytics/AI), and creation of 360-degree view patient
dashboards for full visibility.

6) Cloud gaming

Cloud gaming, a new kind of gaming which streams a live feed of the game directly
to devices, (the game itself is processed and hosted in data centres) is highly
dependent on latency. Cloud gaming companies are looking to build edge servers
as close to gamers as possible in order to reduce latency and provide a fully
responsive and immersive gaming experience.

7) Traffic management

Edge computing can enable more effective city traffic management. Examples of
this include optimising bus frequency given fluctuations in demand, managing the
opening and closing of extra lanes, and, in future, managing autonomous car flows.
With edge computing, there is no need to transport large volumes of traffic data to
the centralised cloud, thus reducing the cost of bandwidth and latency.

8) Content delivery

By caching content – e.g. music, video stream, web pages – at the edge,
improvements to content deliver can be greatly improved. Latency can be reduced
significantly. Content providers are looking to distribute CDNs even more widely
to the edge, thus guaranteeing flexibility and customization on the network
depending on user traffic demands.

Department of Computer Engineering

 An Overview on Edge Computing 23

8. Conclusion…………

Thus, we have successfully studied the technology known as ‘Edge Computing’ by

examining its internal and external working and architecture and understanding the
functions of the various layers present in the system, the functions of each component
covered by the aforementioned layers, technical details and multiple categories of
algorithms which allow us to gain an insight over the working of the computer software
and code, methodology by which the technology is implemented, real world examples
of implementation, and applications of edge computing in multiple fields such as IoT,
aeronautics, automobile industry, gaming, etc.

Department of Computer Engineering

 An Overview on Edge Computing 24

9. Future scope

Following are some of the examples of potential applications of edge computing in

the future -
1) Emergency calls and response before heart attacks
2) Self-driving cars using edge computing to ensure passenger safety
3) Vital signs monitoring and response
4) Non-invasive cancer cell monitoring and response
5) Augmented Reality and Virtual Reality getting a revamp
6) Smart and personalization health nudges (beyond “get up and move”)
7) Electrolyte imbalance monitoring and notification
8) Security upgrades to protect the increasing number of devices connected to the
Internet

Department of Computer Engineering

 An Overview on Edge Computing 25

10. References

10.1 Books
1) Practical Guide to Benchmarking and Experimentation - Nikolaus Hansen
Inria, Research Centre Saclay, CMAP, Ecole polytechnique, Université Paris-
Saclay

10.2 Research papers

[1] Q. Qi and F. Tao, "A Smart Manufacturing Service System Based on Edge
Computing, Fog Computing, and Cloud Computing," in IEEE Access, vol. 7, pp.
86769-86777, 2019, doi: 10.1109/ACCESS.2019.2923610.
[2] Dolui, Koustabh & Datta, Soumya Kanti. (2017). Comparison of edge computing
implementations: Fog computing, cloudlet and mobile edge computing. 1-6.
10.1109/GIOTS.2017.8016213.
[3] Udeen, Haseeb & Abbas, Sadia & Mumtaz, Asia & Hussain, Farya & Kousar,
Kashifa. (2020). Critical Analysis of Edge Computing. 1092-1099.
10.1109/ICSIP49896.2020.9339378.
[4] H. Atlam, R. Walters, and G. Wills, “Fog Computing and the Internet of Things:
A Review,” Big Data Cogn. Comput., vol. 2, no. 2, p. 10, 2018, doi:
10.3390/bdcc2020010.
[5] Cao, Keyan & Liu, Yefan & Meng, Gongjie & Sun, Qimeng. (2020). An Overview
on Edge Computing Research. IEEE Access. PP. 1-1.
10.1109/ACCESS.2020.2991734.
[6] Babar, M., & Sohail Khan, M. (2021). ScalEdge: A framework for scalable edge
computing in Internet of things–based smart systems. International Journal of
Distributed Sensor Networks. https://doi.org/10.1177/15501477211035332
[7] Merenda, M.; Porcaro, C.; Iero, D. Edge Machine Learning for AI-Enabled IoT
Devices: A Review. Sensors 2020, 20, 2533. https://doi.org/10.3390/s20092533
[8] Linna, Ruan & Yan, Yong & Guo, Shaoyong & Wen, Fushuan & Xuesong, Qiu.
(2019). Priority-Based Residential Energy Management With Collaborative Edge and
Cloud Computing. IEEE Transactions on Industrial Informatics. PP. 1-1.
10.1109/TII.2019.2933631.

Department of Computer Engineering

 An Overview on Edge Computing 26

10.3 Internet Link

[1] D. Evans. The Internet of Things How The Next Evolution of the Internet
is Changing Everything. Accessed: Dec. 3, 2016. [Online].
https://www.researchgate.net/publication/30612290

[2] V. Turner, J. F. Gantz, and D. Reinsel. (Nov. 26, 2018). The digital
universe of opportunities: Rich Data and the Increasing Value of the
Internet of Things. [Online].
https://www.emc.com/leadership/digitaluniverse/2014iview/index.html

[3] Cisco Research Center Requests for Proposals (RFPs).

Fog computing, ecosystem, architecture and
applications.
http://www.cisco.com/web/about/ac50/ac207/university/RFP/rfp13078.html

Department of Computer Engineering