![Chapter 1
Introduction of fog computing
Abstract
Purpose: In this chapter, including its history, introductions, benefits, dis-
advantages, applications, and conclusion.
Methodology: A decentralized computing facility known as pervasive computing,
fog networking, or excessive moisture is one in which data, calculations, storage,
and implementations are generated during the most suitable and appropriate place
on-premises data centers and the internet.
Findings: Other names for fog computing include accumulation of dust and fog
networking. The notion of pervasive computing is essentially an extension of
cloud-based solutions and the advantages it delivers to the network’s edges.
Practical implication: This brings the benefits and capabilities of the cloud much
closer to the location where data are produced and responded upon.
1.1 Introduction
Cloud computing that is extended to the perimeter of an organization’s network is
referred to as “fog computing,” which is a phrase that was developed by Cisco. In
certain circles, in addition to the term “edge computing,” it is also known as “per-
vasive computing.” The performance of computation, storage, and communications
services may be facilitated more easily between network elements and the data cen-
ters that handle cloud-based applications thanks to cloud environment. The term “fog
computing,” which is synonymous with “fog networking,” refers to a decentralized
computing environment in which computing, storage, and enterprise applications are
made available in perhaps the most reasonable and realistic place at any moment
along with the continuous spectrum from the data provider to the cloud in Figure 1.1.
Fog computing is also sometimes referred to as edge computing [1].
1.1.1 History of fog computing
The concept of fog computing can be traced back to 2012 when Cisco first intro-
duced the term. They envisioned fog computing as a paradigm that extends cloud
computing to the edge of the network, enabling a seamless connection between
cloud data centers and end devices. The goal was to address the growing demand](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-24-320.jpg)
![for low-latency, high-bandwidth applications in the era of the Internet of Things
(IoT) and other data-intensive technologies.
In the following years, fog computing gained traction as various tech compa-
nies and researchers started exploring its potential. By 2014, the OpenFog
Consortium was formed, a collaborative effort among academic institutions,
industry leaders, and governmental organizations to standardize and promote fog-
computing architectures. This further fueled the development of fog-computing
technologies, leading to their integration into diverse fields, such as smart cities,
healthcare, autonomous vehicles, and industrial automation.
Fog computing has been a hot topic in the IT industry in recent years, and on
November 19, 2015, the OpenFog Consortium was formed. Jeff Faders, Intel’s IoT
Strategist, is the consortium’s first president and Cisco’s Sr. Managing Director
Helder Antunes is its first chairperson [2].
As of 2021, fog computing continues to evolve and solidify its position as a
vital component of the distributed computing landscape, enabling efficient and
responsive data processing at the edge of the network.
1.1.2 Concept of fog computing
● A considerable volume of data is generated by IoT applications. These data
have to be analyzed in order to make interoperability choices and to carry out a
variety of activities.
● The cloud presents a variety of concerns, such as congestion, extremely high-
bandwidth use, delays in real-time answers, and centralized data placement,
when these data are sent.
● Cisco invented the phrase “fog computing” in 2012 in order to address these
issues encountered by wireless sensor networks in the cloud infrastructure.
Connected Server
Sharing Information
Storage Capacity
Provider Cloud Data
Figure 1.1 Fog computing
2 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-25-320.jpg)
![● To minimize latency and maximize spectral efficiency, it proposes to move
processing closer to the end devices.
● The proliferation of sensor-based gadgets has resulted in an enormous volume
of data. Nonvolatile memory and processing are both required for these data.
Difficult, expensive, and time-consuming: preserving data in the cloud. It
reduces operational time and costs by locating capabilities close to the equip-
ment that will be used [3].
1.1.3 What is fog computing
The phrase “fog computing,” invented by Cisco, is often used interchangeably with
“edge computing.” A dense computing framework at the edge of the network has been
referred to as a fog foundation. Such systems are said to include features, such as
reduced latency, reduced latency, and wireless connectivity. Real-time analytics and
enhanced security are among the advantages. A fog-computing infrastructure, on the
other hand, would be able to analyze everything from the network center to the edge of
the network. Using fog computing, a system may adjust its signals depending on traffic
monitoring in order to avoid accidents or minimize traffic congestion. Cloud-based
analytics may also be used to store data for extended periods of time [4].
Cisco’s other examples involve rail safety, smart transmission and distribution
restoration, and information security. In addition to key enabling technologies like
interactive lighting and smart transportation meters, PrismTech Vortex cites
vehicle-to-vehicle and vehicle-to-cloud connectivities. Cisco has provided an
example of the analytics that may be conducted along a fog network in the
picture below.
1.1.4 Why is fog computing
IoT services, such as lower transmission assistance, situational awareness, and geo-
distribution, are among the key goals of fog computing [2]. Many operations that need
minimal delay between IoT devices and the closest fog web service or cloud service for
local data analysis created may benefit from fog computing’s ability to extend cloud
data center capabilities, such as computation, storage, and networking equipment [5].
The number of fog-based applications is increasing. Unconventional applica-
tion framework requires considerable platform features that can only be provided if
the program is compiled close to the end-users if new use cases for the fog envir-
onment are to be realized. All the referenced applications are studied, and their
justifications for employing a fog platform are identified.
Augmented reality games, for example, need a latency of less than 10–20 ms
for end-to-end latencies (connectivity and processor delay included). Propagation
delay from 20 to 40 ms (over communication links) to up to 150 ms (over wireless
networks), the distance between an end-user and the closest cloud data center (over
4G mobile networks). As a result, real-world usage of these apps is impossible.
Run on the internet: Deploying the server element of these apps in fog plat-
forms would be an easy way to lower the overall latency [6]. Video security cam-
eras and other edge devices generate enormous amounts of raw data on a daily
Introduction of fog computing 3](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-26-320.jpg)
![basis, which necessitates optimization of their available bandwidth. Huge network
traffic is generated when so much data are sent to the cloud. Fog computing is an
intermediary in reducing network traffic for these kinds of applications. In order to
preprocess original information at the sources before transferring it to the cloud,
fog middlewares are used.
Smartphones, tablets, and other infrastructure components, such as smart IoT
devices, have a limited amount of computational capacity. High-complexity
applications like image retrieval take a long time to execute on these smart-
phones. Delegation of authority for certain work to the intermediate virtual
machines may affect productivity. When a cloud server is overloaded, the server-
side of the running apps may be offloaded to fog servers. Alternatively, offloading
might be another option.
For many applications, privacy and security are of utmost importance. There is
a large quantity of patient data that may be retrieved via E-health apps in healthcare
administration. To ensure long-term accessibility, most recorded data are stored in
the public cloud. However, many hospitals are concerned about the theft of per-
sonal medical records. By offering storage space to the user or the hospital, private
fog minimizes the problem of data privacy and security.
Many IoT sensors and actuators may need sophisticated computers to run and
control, such as implementation and condition monitoring, system implementation
and switch on/off, and service distribution and incident management. By acting as a
middleman, fog computing may supply computational power that not only makes it
possible to operate various devices but also lets users tailor the services they get to
their own needs [7].
Edge equipment, such as infusion pumps, heartbeat detection systems, and
other monitoring devices, have improved in hospitals. As a result, the US suffers
from the third-leading cause of mortality each year [8] due to difficulties inte-
grating these devices with patients. Remotely hosted apps that interface directly
with the monitoring devices allow for dynamic responses based on real-time data,
thanks to fog computing.
In the IoT context, one of the most pressing concerns is the amount of energy
used by massive IoT devices. In order to save energy, fog computing allows these
appliances to make intelligent choices, such as turning on/off/hibernating.
A pay-as-you-go strategy, which is more common in conventional cloud
computing, reduces costs by charging a flat rate per unit of utilization. One-time
costs for procuring private fog resources may be better than cloud costs for appli-
cations [9]. Reducing network traffic and enhancing response times may be
accomplished in part via the use of content parallelization and networking
protocols.
1.2 How fog computing works
While edge embedded, systems create and obtain information; they lack the com-
putational and storage capacity to execute sophisticated processing and machine
4 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-27-320.jpg)
![learning activities. Despite the fact that cloud servers have the capacity to perform
these things, they are frequently too far away to analyze the data and reply in a
reasonable timeframe. In addition, when maintaining the confidentiality of data
subject to rules in various countries, having all endpoints connect to and transfer
raw information to the server through the internet might have anonymity, con-
fidentiality, and legal concerns [10].
Fog reduces the quantity of data transferred to the cloud by doing processing
on a data hub in a connected home or on a smart router or entry point. Fog net-
working enables short-term predictive modeling at the edge, while the cloud han-
dles the source of energy, long-term business intelligence. Fog networking does not
replace cloud technology.
1.3 Taxonomy of fog computing
Fog computing has suggested classifications, and this is what it looks like.
Categorization of edge computing presents a categorization of the current fog-
computing efforts. Network architecture is highlighted by the categorization in
the following ways.
Configuration of fog nodes: At the edge of the network, the nodes with diverse
architectures and settings may support fog-computing infrastructure.
● Node-to-node communication: An edge network’s methods for coordinating
nodal cooperation among several fog nodes.
● Metric for resource/service provisioning: How to provide resources and ser-
vices in a cost-effective manner under a wide range of conditions.
● Goals for the quality service.
By introducing edge computing as a middle layer among cloud applications and end
applications, the Service Level Objectives (SLOs) were achieved. A network system
is appropriate for the situation. Fog computing is an extension of previous computing
paradigms that have been implemented in various networking systems. Security is a
major issue. Fog computing’s security considerations vary depending on the situa-
tion. Numerous taxonomic groupings are covered by existing methods and solutions.
According to fog-computing features, this taxonomy does not accurately reflect the
relative effectiveness of all of the many recommended approaches. Different
execution settings, networking topologies, application characteristics, resource
architecture, and so on are all taken into consideration and addressed in the work that
has been viewed in this study. Due to the complexity of fog computing, it is almost
impossible to pinpoint the most optimal solution in terms of structure, service, and
security. In the existing studies, fog nodes have been categorized into five different
types: servers, networking devices, cloudlets, base stations, and automobiles. In-
house web servers may be seen at bus terminals, shopping centers, roads, and even in
public parks. In the same way that being light in weight is similar to because of
virtualization and cloud computing, these fog servers are cloud servers. The fog
server is one of fog computing’s most critical functional properties. Fog servers are
Introduction of fog computing 5](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-28-320.jpg)
![referred to as microservices, micro data centers, nano servers, and so on in certain
studies based on their physical size, while they are classified as cache servers, cal-
culation servers, storage servers, and so on in other papers based on their functions.
Fog computing may benefit from a server-based node design that increases com-
puting and storage capability. However, it restricts the extent to which the execution
environment may be used [11].
● Devices for connecting to the internet: It is feasible that Fog-computing
infrastructures might be built using devices like routers, switches, and set-top
boxes, in addition to their typical networking duties. Several modern switches
and routers include a variety of system resources, such as CPUs, extensible
main and secondary memory, and programing platforms. In addition to stan-
dard hardware and software components, certain specialized network interface
cards, such as intelligent gateways and IoT hubs, have been represented as fog
nodes in other publications. Network devices deployed in a dispersed way
increase the prevalence of pervasive computing, but the physical diversity of
devices has a significant influence on the provisioning of services and
resources. The cloudlet is a micro-cloud that resides in the middle of the end
device, cloudlet, and cloud hierarchy. Cloudlets may be used to complement
MCC by providing mobile device users with cloud-based services. According
to a series of studies, cloudlets have been referred to as “Fog nodes.” A huge
number of end devices may be handled concurrently using cloudlet-based fog
computing. Cloudlets, although being deployed at the edge, may nonetheless
function as centralized components in certain circumstances owing to struc-
tural restrictions. There are still serious issues with fog computing that prevent
it from supporting IoT [12].
● Stations at the ground zero: An essential part of any wireless network, base
stations process and transmit data to and from the mobile nodes. Traditional
base stations equipped with particular storage and computation capabilities
have recently been deemed viable for fog computing in recent research. Fog
nodes may be created using RSUs, small cell access points, and so on, much as
regular base stations. Fog-based extensions of cloud radio access network
(CRAN), vehicular ad hoc network, and similar networks are better served by
base stations. Fog creation using base stations, on the other hand, is compli-
cated by high deployment costs and networking interference.
● Intelligent vehicles: Fog nodes may be placed in moving or stationary cars at
the edge of a network with computing resources. It is possible to create a
highly dispersed and scalable fog environment using vehicles. It will, however,
be very difficult to provide privacy and fault tolerance while maintaining
optimal quality of service (QoS) in such an environment.
● Collaboration at the node level: Cooperative approaches for cluster, peer-to-
peer (P2P), and master/slave computation nodes for fog have all been reported in
several studies. Cluster nodes in the network may form their own clusters in
addition to maintaining a collaborative computing environment. When fog nodes
are in close proximity to one another, they might form clusters. The creation of
6 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-29-320.jpg)
![functional subsystems and congestion control may be prioritized while the nodes
are forming a cluster. It is possible to use the capabilities of several fog nodes
concurrently by using cluster-based cooperation. It’s tough to scale static clus-
ters at runtime because dynamic cluster generation is heavily reliant on the
current demand and readily available fog nodes. In both circumstances, the
networking overhead is critical to the overall outcome. P2P cooperation is highly
popular in fog computing because of the distributed nature of the system. It is
possible to do P2P cooperation in a hierarchical or flat manner. There are a
variety of ways to classify P2P cooperation between fog nodes besides proxi-
mity. For example, virtual computer instances are shared among nodes in a
cloud computing environment to optimize resource utilization and deliver scal-
able and cost-effective solutions to users. It is about P2P cooperation rather than
relying on a single node’s processed output alone. In P2P cooperation, fog nodes
may be easily augmented and made reusable. P2P nodal cooperation, on the
other hand, is plagued by problems with dependability and access control [13].
● Master–slave: Master–slave nodal cooperation has been extensively discussed
in a number of publications. A master fog node typically manages slave nodes’
functioning, processing load, environmental protection, information flows, and
other aspects. The fog-computing environment may also create a hybrid col-
laborative network using a master–slave strategy, cluster nodal connections,
and P2P interactions. As a result, both the master and the slave fog nodes in
order to process data in real time require high-bandwidth communication.
There are several aspects that play a role in the supply of resources and ser-
vices in fog computing, time, energy, application and database context, and
more are all taken into account.
● Time: Fog computing makes efficient use of time as a key consideration in
resource and service supply. The amount of time it takes to complete a job is
known as computation time. It’s important to keep in mind that the amount of
time it takes to operate an application relies heavily on the configuration of the
resources it’s operating on. In addition, the time it takes to compute a task helps
to distinguish between the current and previous periods of different programs and
has a substantial impact on fog’s resource and power management. The time it
takes for data items to be exchanged in a fog-computing environment is referred
to as communication time. It has been explored in two ways in the literature:
There is a direct connection between the end devices/sensors and fog nodes. To
aid in task execution, the network context is reflected in the required commu-
nication time. A system’s deadline sets the maximum amount of time it may go
without receiving a service. Task completion satisfaction has been regarded as a
significant QoS indicator in various studies. The delivering services constraint
distinguishes between latency sensitivity and latency tolerance applications and
games. Service access times in a multitenant cloud architecture, service reaction
times, and other time-based metrics such as these may be studied for efficient
function and resource deployment and management in sensor networks [14].
● Data: Fog-computing literature makes frequent use of input data size and data
flow characteristics, two data-centric metrics. The quantity of data that must be
Introduction of fog computing 7](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-30-320.jpg)
![processed via fog computing is referred to as data size. There have been sev-
eral discussions about the computational space needs of requests in relation to
data size. In addition, data gathered from a large number of dispersed sensors
and devices may have the characteristics of Big Data. Provisioning resources
and services based on data load might be a useful strategy in this situation.
Data size also has a significant effect on determining whether a computing
activity should be performed locally or remotely. The properties of data
transmission are defined by the data flow. Event-driven or real-time data flow
in the fog-computing environment may have a significant impact on resource
and service delivery. In addition, abrupt changes in data flow might lead to
dynamic node load balancing. Fog computing’s resource and service provi-
sioning may also be analyzed in terms of heterogeneous data architecture, data
semantic norms, and data integrity needs [15].
● Cost: Fog resource and service delivery may be heavily influenced by con-
siderations relating to cost, both from the standpoint of service providers and
customers. It is easy to see how broadband use and associated expenses have a
direct impact on connectivity costs in a fog-computing system. Some studies
attribute connectivity prices to the upload of data from end devices/sensors and
the exchange of data across nodes, while others attribute networking delays
caused by broadband problems to network security costs. In a fog-computing
environment, deployment costs are mostly tied to the costs of setting up the
infrastructure. In several studies, efficient resource and service supply has been
linked to cost-effective infrastructure implementation. The cost of deploying
infrastructure may be broken down into two parts: the actual deployment of fog
nodes in the network, and the creation of virtual computing instances inside
those nodes. Fog nodes’ computational costs when executing applications or
processing activities are referred to as “execution costs.” The use of execution
costs in resource provisioning and invoicing is rare in fog computing despite its
widespread usage in other computing paradigms. Task completion time and
resource utilization costs have been used to compute the overall cost of these
tasks. Fog-computing resource and service provisioning may take into account
migration costs as well as the previously listed costs, as well as charges for
security precautions, the most a customer is willing to pay for a product or
service [16].
● Consumption of energy and environmental impact: In a few studies, fog
resources and services have been prioritized based on energy concerns. Fog-
cloud interaction has been studied extensively for its energy consumption
across all devices, as well as the trade-off between energy and delay at various
stages of the process. Previously, the carbon emission rate per unit energy
consumption of various nodes was taken into account for resource provisioning
objectives in another study. Fog resources may be provisioned according to the
energy constraints of end devices/sensors, such as residual battery life and the
energy characteristics of communication media. Context refers to the condi-
tions under which a certain thing is found. For resource and service delivery,
user and application context has been studied in fog-based research papers. In
8 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-31-320.jpg)
![the future, resources may be allocated to a user based on the user’s attributes
(e.g., service use history and service relinquish likelihood). Service and
resource provisioning may benefit from customer input, such as net promoter
score and customer needs. Service provisioning in previous works has taken into
account user density, mobility, and network state. The context of an application
may be defined as the operational needs of several applications. Prerequisites for
task performance (such as processor speed, storage, and storage), as well as
network connectivity, and other operational needs might impact the supply of
resources and services. The present workload of various apps has also been taken
into account as an application context in other research. Fog-computing contexts
may also be considered in terms of the execution environment, node character-
istics, application design, and so on, and these contexts can play a significant role
in providing resources and services. As a result, every piece of background
material must be thoroughly examined [17].
● Goals for the quality of service: Certain SLOs have been suggested to be
achieved using a variety of application platforms, computational models, and
optimizations of fog node architecture approach in current research. Almost all
of the successfully achieved SLOs are management-oriented and deal with
concerns such as latency and power consumption as well as cost, resource
allocation, data storage, and other types of applications.
● Latency control: Fog computing’s latency control essentially prevents the
eventual service delivery time from exceeding a predetermined threshold for
acceptable latency. A service request’s maximum acceptable latency or an
application’s QoS requirement may fall within this barrier. Some efforts have
placed an emphasis on efficient nodal cooperation start to guarantee that com-
pute activities carried out by collaborated nodes may be completed within the
latency limitation set. It has also been shown that distributing computing tasks
across clients and fog nodes may reduce service request computation and com-
munication delay. In addition, a low-latency fog network design was presented
in another paper to control latency. The primary goal of this project is to find a
node in the fog network that delivers services with the least amount of delay.
● Controlling the costs: Operating expenses (Operating Expenses) may be
evaluated in terms of fog-computing cost management (OPEX). Distribution
of fog nodes and their networks is the primary cause of CAPEX in fog com-
puting. Fog computing’s CAPEX may be kept to a minimum by strategically
placing and using an optimal number of fog nodes. According to this concept,
the total cost of fog computing is reduced by optimizing the positioning and
number of sensor nodes in use. Fog nodes are also referred to as virtual
machines launch vehicles and virtual machines in another study. The cost of
running data processing processes on these virtual machines varies from pro-
vider to provider, and the cost is not always the same. To reduce OPEX in fog
computing, it is possible to take advantage of the cost diversity of fog nodes/
providers. According to this, the article proposes a technique to discover the
best fog nodes to host Virtual Machine (VMs) in order to reduce OPEX in fog
computing [18].
Introduction of fog computing 9](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-32-320.jpg)
![● Management of the network: Fog computing’s network management com-
prises, for example, the control of network congestion in the core, the support
of software-defined networking/network function virtualization (SDN/NFV),
and the guarantee of smooth connection. Network congestion is primarily
caused by an increase in network overhead. Because IoT devices/sensors are
widely dispersed over the edge, the cost on the core network may be greatly
increased by concurrent interactions between end components and cloud data
centers. As a result, network congestion will arise, lowering the system’s
overall performance. In light of this, a layered fog node design has been sug-
gested that allows for the processing of service requests at the node level. As a
result, despite getting large numbers of service requests, clouds only receive
compressed versions of such requests, which have less impact on the network.
There is a lot of interest in virtualizing the traditional networking infra-
structure. Virtualized networks are made possible in large part by SDN. SDN is
an SDN approach that separates control and data planes from communication
gear and puts it into software on different servers. Support for NFV is a crucial
feature of SDN. To put it simply, NFV is an architectural idea that enables
conventional networking tasks to be virtualized so that they may be done via
software. SDN and NFV have a significant impact on cloud-based environ-
ments because of their large variety of services. New network topologies for
fog computing have been developed to allow SDN and NFV as a result of this;
as a result of their physical variety, end devices are able to communicate
seamlessly with other entities such as the cloud or fog or desktop computers or
mobile devices. As a result, finding resources and keeping the network’s
communication and computing capacities up to date are made simpler. It is a
problem that has already been addressed in fog computing, with new archi-
tectures for fog nodes like the IoT hub and fog networking like the vehicular
fog computing (VFC). Another development in fog computing is a policy-
driven framework for ensuring secure connectivity between devices [19].
● Management of computations: Fog computing’s SLOs include a high priority on
ensuring that computational resources are properly managed. It is possible to esti-
mate computer resources, distribute workloads, coordinate computing resources, and
more with fog computing. Resources may be assigned according to certain rules in
fog computing so that suitable resources can be allocated, desired QoS can be
attained, and an exact service fee can be enforced. According to the current litera-
ture, resource estimation strategies are built based on user characteristics experi-
enced QoE, features of service-accessing devices, and so on. Fog computing’s
workload distribution should aim to optimize resource usage while minimizing
computational idle time. More specifically, a balanced load is ensured on several
components. A scheduling-based workload allocation strategy has been imple-
mented in a fog-based research project in order to distribute the computational
burden across fog nodes and client devices. As a result, both parties’ overhead costs
are reduced, which raises QoE. In a separate study, a framework for balancing fog-
cloud communication delays and power consumption was proposed. Because of
their heterogeneity and resource limitations, coordination among various fog
10 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-33-320.jpg)
![resources is absolutely essential. With fog computing, large-scale applications may
be distributed over several fog nodes due to its decentralized nature. Without ade-
quate coordination of fog resources, it would be difficult to achieve the required
performance under these situations. In light of this, a paradigm for managing fog
resources based on directed graph coordination has been developed.
● Management of application programs: Efficient programing platforms are
critical for successful fog-computing application administration. In addition to
the scalability and compute offloading capabilities, these features aid application
administration. Development, compilation, and execution of programs are all
made easier with the help of a programing platform, which includes all of the
components listed above. Due to fog computing’s dynamic nature, ensuring
effective resource management and real-time decision-making capabilities are
paramount for its successful implementation in applications such as IoT and
edge computing. It’s quite tough to develop software to handle large-scale
applications. Mobile fog, a new development platform, has addressed this issue.
Mobile fog’s reduced abstractions of programing paradigms make it feasible to
build large-scale decentralized applications. As well as the coordination of
resources during the implementation, an implementation framework for fog
computing was also established in another document. To keep their QoS high
even as the number of app users increases and unexpected occurrences occur,
apps must be capable of adapting. Application scheduling and service access for
users may both benefit from scaling strategies. Fog computing has recently
presented an architecture for a QoS-aware self-adaptive scheduler to facilitate
the scalable scheduling of data stream applications. Using this scheduler, pro-
grams may be scaled up and down based on the number of users and the amount
of resources available. It is also simpler to arrange programs in a dispersed form
because of the scheduler’s self-adaptive capabilities. Fog computing has also
offered an adaptive approach for users to choose their service access mode
depending on the distance, location, and QoS needs of the service-accessing
entities. It is possible to transmit computational activities from resource-
constrained end devices to more resource-rich devices via offloading methods.
In a mobile cloud environment, computational offloading is a typical occurrence.
As part of fog computing’s compatibility upgrade support for mobile applica-
tions’ computation dynamic provisioning in other communication networks has
been stressed in various publications lately. Mobile apps’ distributed computing
and the availability of resources have been examined in these studies [8].
● Management of data: Fog computing can’t function properly if its informa-
tion systems SLOs aren’t in place. Different research studies have looked at
fog computing’s data management takes a multifaceted approach. Fog com-
puting’s information management approach places a high value on computa-
tional intelligence services and preparatory distribution of resources. As an
alternative, low-bandwidth aggregation from scattered end devices/sensors
may be explored in the interest of improved data administration. End devices/
sensors, on the other hand, have limited storage capacity. Storage enhancement
for end-entity data storage and data may have a substantial influence on fog
Introduction of fog computing 11](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-34-320.jpg)
![computing in this case. Data processing in fog computing has also been
emphasized as a crucial part of storage growth for smart applications.
● Management of energy resources: Power management may be provided as a
service to various fog computing is being used to connect systems. A cloud
infrastructure for fog computing might enable home-based distributed systems
to regulate power with customized user control, according to a report.
Centralized cloud data centers’ power usage may be managed in certain cir-
cumstances using cloud services. Data center power consumption is mainly
dependent on the sort of applications being executed. By offering infra-
structure to support a number of energy-hungry applications, fog computing
may complement cloud data centers in this situation. Therefore, cloud data
centers will be able to maintain proper power management as a result of
reduced energy consumption. Furthermore, carbon footprint emissions can be
lowered by carefully controlling the power used in fog computing.
● Usefulness of networking systems: The IoT relies heavily on fog computing.
Fog computing’s usefulness in different mobility, content management, radio
access, and vehicle networks are all examples of communication networks that
have been emphasized in recent research publications.
● Internet of Things: Every device in the IoT can communicate with each other
and exchange data. The IoT environment can be viewed from many different
angles. Additionally, in various fog-based research studies, this contact has
been categorized as either industrial or home-based execution environments.
Furthermore, fog-computing systems and service models have also taken into
account many kinds of IoT, including networks of wirelessly sensing devices
and cyber–physical systems.
● Access networks for mobile devices/mobile phones: Fog computing’s
applicability to mobile networks has also been studied in a number of studies.
Fog computing’s interoperability with 5G mobile networking has been a major
focus of these studies. Compared to current cellular networks, 5G allows for
substantially faster connectivity, more signal capacity, and reduced latency in
service delivery. Fog computing may be used in various mobile networks other
than 5G, such as 3G and 4G. A different study looked at how workloads in fog
cloud for mobile communication are allocated depending on trade-offs
between power and delay. Individual devices communicate with other net-
work entities via radio connections in the radio access network. CRAN, the
cloud-assisted Radio Access Network (RAN), has already piqued the interest
of many researchers. Fog computing–based radio access networks have also
been investigated as a possible supplement to the capabilities of CRAN [20].
● Power line communications via passive optical network: With the introduction
of LRPON, backhaul services for homes, businesses, and wireless networks can
now take advantage of a new, low-latency, high-bandwidth technology for long-
distance communications. LRPONs facilitate network consolidation in addition to
providing a wide coverage area. Fog computing has been integrated with LRPONs
in order to optimize the network design in this article. Communication over the
smart grid’s power lines (PLC) is common. Data and alternating current are sent
12 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-35-320.jpg)
![concurrently in PLC utilizing electrical connections. Discussion of fog-computing
PLCs in electrical power distribution has been extensive.
● Internet-based platform for the distribution of content: Distributed proxy
servers supply material and provide good reliability and availability for the
end-users via a content distribution network. Many fog-based research projects
use fog nodes as they open up opportunities to make content dissemination
easier. Users may access fog-based content services with little latency since
fog nodes are scattered over the network’s edge. The dissemination of high-
quality information will run more smoothly as a result.
● Network of automobiles: Data interchange and resource augmentation are
made possible by vehicular networks, which allow the autonomous establish-
ment of a wireless communication network among cars. Computational and
networking resources are made available to cars as part of this network. Fog
nodes are vehicles that reside at the edge of a network and are used to promote
a fog computing–based vehicular network in several studies [21].
● Fear for one’s safety: Fog computing relies on an underlying network
between end devices and cloud data centers; it has a high level of security risk.
However, security considerations in cloud computing are essential to safeguard
sensitive data and protect against potential threats or breaches, making robust
encryption, access controls, and continuous monitoring critical components of
a secure cloud environment. In the literature, fog computing has been resear-
ched in terms of information identification, confidentiality, secure data
exchange, denial-of-service (DoS) attack, and so on.
● Authentication: In fog-based systems, user authentication plays a critical role in
preventing infiltration. Unwanted access to fog services is very intolerable since
they are utilized on a “pay-as-you-go” basis. In addition to user authentication, the
secure fog-computing environment has seen device authentication, data transfer
authentication, and instance authentication. End device/sensor data are processed
via privacy fog computing. Sometimes, these statistics are discovered to have a
strong correlation with users’ personal and professional circumstances. As a result,
one of the most pressing issues in fog computing is the protection of user security.
It has been noted that privacy is an issue with fog-based vehicular computing.
● Data encryption: Fog computing is a useful adjunct to cloud computing.
Cloud computing is required in certain circumstances for data handled via fog
computing. Fog nodes must encrypt this data since they frequently include
important information. In light of this, the proposed fog node design includes a
data encryption layer.
● DoS attack: Because fog nodes have a limited amount of resources, they are
unable to accommodate a high number of simultaneous queries. Fog comput-
ing’s performance may suffer greatly in this situation. DoS attacks might be
crucial in causing such significant service outages in fog computing. Fog nodes
may be kept busy for a longer length of time by concurrently issuing a large
number of irrelevant service requests. Due to the lack of resources, helpful
services are no longer accessible. Fog computing has been used to discuss and
clarify this kind of DoS attack.
Introduction of fog computing 13](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-36-320.jpg)
![● Inquiry into the gap and possible future directions: Rather than relying on
remote servers, fog computing makes advantage of cloud resources that are
already nearby. Fog computing is critical for supporting widely dispersed end
devices and sensors. As a result, fog computing has emerged as a major
research area in both academia and industry in recent years. There is an
overview of several reviewed publications on fog computing. Many important
aspects of fog computing have been discussed already, but there are still some
issues that need to be addressed if this field is to advance further. In this
section, we will talk about some of the gaps in the existing literature and
possible future research directions [22].
Provisioning of resources and services in light of current situation: Fog comput-
ing’s resource and service provisioning may benefit from context awareness. Fog
computing may acquire contextual information in a variety of ways, such as
● Location, time (peak, off-peak), and so on.
● The application’s perspective: latency sensitivities, application design, and
so on.
● Mobility, social connections, activity, and so on are all examples of user
context.
What resources are accessible on the device? How much juice is left in the
battery?
The context of a network might include factors, such as bandwidth and traffic.
There are still many undiscovered features of background information, ignor-
ing the fact that several fog-based research studies have taken into account specific
information while assessing how resources and service operations might be studied
using fog-based research methods [23].
1.4 Fog computing versus cloud computing
Many people use the words fog computing and edge computing alternatively
because they both entail moving intelligence and processors closer to the location
where data are produced. However, the primary distinction between the two is the
location of intellect and computing capacity.
● Cloud technology: The processing of data and applications in the cloud is
time-consuming for huge datasets. Bandwidth issues result from the trans-
mission of all data over cloud channels. Because of distant servers, slow
response times and scalability issues arise.
● Cloud computing: Instead of displaying and operating from a central cloud,
fog operates at the network’s periphery. Therefore, it uses less time. Less need
for bandwidth, since all data are pooled at a single access point as opposed to
being sent through cloud channels. It is conceivable for a fog-computing
platform to circumvent reaction time and scalability difficulties by selling tiny
machines known as edge servers in direct user view.
14 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-37-320.jpg)
![1.5 Fog computing and IoT
Do not be astonished to learn that about 31 billion IoT devices are now in operation. It
is no surprise that we generate 2.5 quintillion bytes of data every day. Clearly, we
need advantages over conventional techniques of data management. This is when fog
computing comes into play. When an application or device accumulates huge
amounts of data, effective data warehousing becomes difficult, not to mention
expensive and difficult [24]. Heavy data place a strain on bandwidth usage. It is costly
to build Big Data centers to store and arrange this data. Fog computing collects and
distributes storage, computation, and network connection services, decreases energy
consumption, increasing the productivity and value of the data, and decreases space
and temporal complexity. Consider two IoT examples:
● Sustainable urban: Data centers are not designed to accommodate the rising
demand for smart city applications. As more individuals began to use IoT
devices, more information would be transferred and accessible. Fog computing
may assist such inadequate facilities and smart grids in delivering the true
benefit of IoT application development.
● Facilities: The word “utilities” encompasses applications like hospitals, rail-
roads, and enforcement agencies that need the most advanced data delivery
infrastructure to enhance their functioning. Knowledge on energy consump-
tion, fractures, and water leaks, for instance, may be utilized to update payment
information, save lives, and enhance operations. How cloud technology
increases the value of solutions for the IoT and end-users is getting more
potent. A significant proportion of data is currently processed in the cloud in
Figure 1.2. In addition, here are six advantages that fog computing may pro-
vide to the IoT design and development phase.
● Maximum business flexibility: With the proper tools, fog apps may be
developed and deployed as required. These programs allow the user to custo-
mize the device’s behavior.
● Enhanced security: As a proxy for devices with limited resources, fog com-
puting updates their software and security credentials. It installs fog nodes with
Huge Business
Agility
Better Security
System
Low-latency Rate
Network
Bandwidth
Efficiency
No Interruption
Services
Refined User
Experience
Figure 1.2 Fog computing enhances solution
Introduction of fog computing 15](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-38-320.jpg)
![1.5.1 Are fog computing and edge computing the same?
This is a challenging question. To keep things simple, fog and edge computing are
virtually the same. Both methods employ computational performance to bring
intelligence back to the lowest area network level of the network architecture.
This inhibits the execution of computing activities in the cloud, saving time,
resources, and money. In addition, both cloud technology and edge computing may
help organizations lessen their dependency on cloud-based platforms for data
gathering, which lowers latency difficulties and the time required to make data-
driven choices [25].
Perhaps one of the most notable distinctions is the data processing. With a fog
node, data are packaged in fog. Computing at the edge interprets information in the
system or sensor without transferring it to another infrastructure.
With effective collecting and analyzing data in real time, however, both technol-
ogies save time and money when it comes to sustaining procedures. Imagine receiving
statistics in near real time that are valuable for improving performance and enhancing
uptime. Both fog computing and networking technologies make this feasible.
● Fog computing and IoT app development in action: Fog computing stands
out as a reliable, dynamic, and cutting-edge technology in a wide range of
fields. In this part, we will look at four real-time examples: On the fog plat-
form, data transfer for video streaming applications is well organized. Because
of the flexibility and scalability of fog networking and real-time data proces-
sing, this is possible. In addition, fog encourages interaction in a virtual full-
featured system, allowing real-time video business intelligence for security
cameras [26].
● Monitor and control systems in the healthcare sector: Future healthcare
choices would be incomplete without comprehensive and real-time health data.
Data transfer in real time is feasible; however, thanks to the implementation of
fog-computing frameworks. “U-Fall” is another key use case, since it auto-
matically identifies a large fall in the event of mild strokes.
● Playing video games: Fog computing, like the cloud, puts computational
power closer to the players’ fingertips. It is no secret that SEGA’s fog gaming
system relies on the low latency provided by local gaming arcades and centers.
So, instead of streaming from the cloud, players would use the local arcade
equipment’s CPUs to power up their games. The dispersed devices improve the
quality of the online gaming experience for many players: a system for intel-
ligent traffic light control; imagine a traffic signal system that is smart uses fog
nodes. Multiple sensors on the node interact locally to detect the presence of
bicyclists or walkers, as well as the speed and distance traveled by cars. The
green light sends out warnings based on the information. Since it already
monitors video security cameras, an ambulance may be easily seen by its
emergency light and warning alarm. Allowing a vehicle to pass traffic may be
done by adjusting the traffic signals in the area [27].
● In a nutshell: The IoT solution generates enormous amounts of data every
day, and fog computing serves as a partner to the cloud. Data processing near
Introduction of fog computing 17](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-40-320.jpg)
![● Identification and privacy considerations.
● Information infrastructure for wireless connections.
1.11 Application of fog computing
The IoT makes extensive use of the developing technology of fog computing. The
network edge receives data and services from the network core via fog computing.
Similar to the cloud, the fog provides data, computation, storage, and enterprise
applications to end-users.
● Patient monitoring systems in real-time primary healthcare units are being
developed.
● Monitoring of pipelines for leaks, fires, theft, and the like.
● Smart grid control has the ability to switch between various sources of energy.
● Smart farms with agricultural applications and irrigation management systems
are being used in agriculture.
● Fleet management and vehicle health monitoring equipment for trucks and
buses are also included in this category.
● Automated invoicing and reporting of shopping baskets.
● Fire alarms, temperature control, and intrusion detection are all examples of
smart home technology.
1.12 Conclusion
Fog-computing applications come in many shapes and sizes, and as a result, the
pervasive computing platforms built to serve them must be able to handle needs
that span a broad spectrum. Researchers hope that these findings will help future
fog platform developers make well-informed judgments regarding the capabilities.
They may or may not include the kinds of activities that would most benefit from
those inclusions. Fogging is a term used to describe a hardware implementation
where data, processing, and applications are centered at the edge of the network
instead of being on the internet. Service latencies and customer satisfaction are both
improved consequently. Fog computing eliminates the need for a return trip to the
cloud for analysis, allowing users to react to events faster. This leads to increased
organizational capabilities, better service levels, as well as greater safety for
enterprises that use fog computing.
References
[1] S. Yi, C. Li, and Q. Li, A Survey of Fog Computing: Concepts, Applications
and Issues, In Proceedings of the 2015 Workshop on Mobile Big Data, New
York, USA, 2015, pp. 37–42.
Introduction of fog computing 27](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-50-320.jpg)
![[2] F. Bonomi, R. Milito, J. Zhu, and S. Addepalli, Fog Computing and Its Role
in the Internet of Things, In Proceedings of the First Edition of the MCC
Workshop on Mobile Cloud Computing, New York, USA, 2012, pp. 13–16.
[3] S. Yangui, P. Ravindran, O. Bibani et al., A Platform As-a-Service for
Hybrid Cloud/Fog Environments, In IEEE International Symposium on
Local and Metropolitan Area Networks (LANMAN), 2016, pp. 1–7.
[4] M. Aazam and E. N. Huh, Dynamic Resource Provisioning through Fog
Micro Datacenter, In IEEE International Conference on Pervasive
Computing and Communication Workshops (PerCom Workshops), 2015,
pp. 105–110.
[5] O. Salman, I. Elhajj, A. Kayssi, and A. Chehab, Edge Computing Enabling
the Internet of Things, In IEEE 2nd World Forum on Internet of Things (WF-
IoT), 2015, pp. 603–608.
[6] H. T. Dinh, C. Lee, D. Niyato, and P. Wang, A Survey of Mobile Cloud
Computing: Architecture, Applications, and Approaches, Wireless
Communications and Mobile Computing, 13, 1587–1611, 2013.
[7] J. M. Kang, H. Bannazadeh, H. Rahimi, T. Lin, M. Faraji, and A.
LeonGarcia, Software-defined infrastructure and the Future Central Office,
In IEEE International Conference on Communications Workshops (ICC),
2013, pp. 225–229.
[8] H. Hejazi, H. Rajab, T. Cinkler, and L. Lengyel, Survey of Platforms for
Massive IoT, In IEEE International Conference on Future IoT Technologies
(Future IoT), 2018, pp. 1–8.
[9] W. Shi, J. Cao, Q. Zhang, Y. Li, and L. Xu, Edge Computing: Vision and
Challenges, IEEE Internet of Things Journal, 3, 637–646, 2016.
[10] F. Bonomi, R. Milito, P. Natarajan, and J. Zhu, Fog Computing: A Platform
for Internet of Things and Analytics in Big Data and Internet of Things: A
Roadmap for Smart Environments, Springer, Cham, 2014, pp. 169–186.
[11] S. Yi, Z. Qin, and Q. Li, Security and Privacy Issues of Fog Computing: A
Survey, In Proceedings of the International Conference on Wireless
Algorithms, Systems, and Applications, 2015, pp. 685–695.
[12] P. Hu, S. Dhelim, H. Ning, and T. Qiu, Survey on Fog Computing:
Architecture, Key Technologies, Applications and Open Issues, Journal of
Network and Computer Applications, 98, 27–42, 2017.
[13] A. V. Dastjerdi, H. Gupta, R. N. Calheiros, S. K. Ghosh, and R. Buyya, Fog
Computing: Principles, Architectures, and Applications ArXiv160102752
Cs, 2016.
[14] A. V. Dastjerdi and R. Buyya, Fog Computing: Helping the Internet of
Things Realize Its Potential, Computer, 49, 112–116, 2016.
[15] C. C. Byers, Architectural Imperatives for Fog Computing: Use Cases,
Requirements, and Architectural Techniques for Fog-Enabled IoT Networks,
IEEE Communications Magazine, 55, pp. 14–20, 2017.
[16] R. Mahmud, R. Kotagiri, and R. Buyya, Fog Computing: A Taxonomy,
Survey and Future Directions in Internet of Everything, Springer, Singapore,
2018, pp. 103–130.
28 Enabling technologies for smart fog computing](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-51-320.jpg)
![[17] C. Mouradian, D. Naboulsi, S. Yangui, R. H. Glitho, M. J. Morrow, and P. A.
Polakos, A Comprehensive Survey on Fog Computing: State-of-the-Art and
Research Challenges, IEEE Communications Surveys Tutorials, 20, 416–
464, 2018.
[18] R. K. Naha, S. Garg, and A. Chan, Fog Computing Architecture: Survey and
Challenges, arXiv:1811.09047 [cs], Nov. 2018.
[19] A. Yousefpour, C. Fung, T. Nguyen et al., All One Needs to Know about
Fog Computing and Related Edge Computing Paradigms: A Complete
Survey, Journal of Systems Architecture, 98, 289–330, 2019.
[20] M. Eder, Hypervisor-vs. Container-based Virtualization in Future Internet
(FI) and Innovative Internet Technologies and Mobile Communications
(IITM), Seminars FI / IITM WS 15/16, Network Architectures and Services,
1, pp. 11–17, 2016.
[21] D. Zeng, L. Gu, S. Guo, Z. Cheng, and S. Yu, Joint Optimization of Task
Scheduling and Image Placement in Fog Computing Sup-ported Software
Defined Embedded System, IEEE Transactions on Computers, 65,
pp. 3702–3712, 2016.
[22] H. Xiang, W. Zhou, M. Daneshmand, and M. Peng, Network Slicing in Fog
Radio Access Networks: Issues and Challenges, IEEE Communications
Magazine, 55, pp. 110–116, 2017.
[23] J. Santos, T. Walters, B. Volckaeert and F. De Turck, Fog Computing:
Enabling the Management and Orchestration of Smart City Applications in
5G Networks, MDPI Journal Entropy, 2017, 6, pp. 4–15.
[24] D. Zhao, D. Liao, G. Sun, and S. Xu, Towards Resource-Efficient Service
Function Chain Deployment in Cloud-Fog Computing, IEEE Access, 6,
pp. 66754–66766, 2018.
[25] A. Shawish and M. Salama, Cloud Computing: Paradigms and Technologies
in Inter-Cooperative Collective Intelligence: Techniques and Applications,
Springer, Berlin, Heidelberg, 2014, pp. 39–67.
[26] M. Armbrust, A. Fox, R. Griffith, et al., A View of Cloud Computing
Commun ACM, 53, 50-58, 2010. P. Mell and T. Grance, The NIST
Definition of Cloud Computing National Institute of Standards and
Technology and Information Technology Laboratory, 2009.
[27] R. Jain and S. Paul, Network Virtualization and Software Defined
Networking for Cloud Computing: A Survey, IEEE Communications
Magazine, 51, 24–31, 2013.
Introduction of fog computing 29](https://image.slidesharecdn.com/27432221-250509143222-d5f5f968/85/Enabling-Technologies-For-Smart-Fog-Computing-Kuldeep-Singh-Kaswan-52-320.jpg)
Enabling Technologies For Smart Fog Computing Kuldeep Singh Kaswan
- 1. Enabling Technologies For Smart Fog Computing Kuldeep Singh Kaswan download https://ebookbell.com/product/enabling-technologies-for-smart- fog-computing-kuldeep-singh-kaswan-54864442 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Toward Social Internet Of Things Siot Enabling Technologies Architectures And Applications Emerging Technologies For Connected And Smart Social Objects 1st Ed Aboul Ella Hassanien https://ebookbell.com/product/toward-social-internet-of-things-siot- enabling-technologies-architectures-and-applications-emerging- technologies-for-connected-and-smart-social-objects-1st-ed-aboul-ella- hassanien-10489162 Enabling Technologies For Effective Planning And Management In Sustainable Smart Cities Mohd Abdul Ahad https://ebookbell.com/product/enabling-technologies-for-effective- planning-and-management-in-sustainable-smart-cities-mohd-abdul- ahad-49406944 Enabling Technologies For Social Distancing Fundamentals Concepts And Solutions Diep N Nguyen https://ebookbell.com/product/enabling-technologies-for-social- distancing-fundamentals-concepts-and-solutions-diep-n-nguyen-44888200 Enabling Technologies For Next Generation Wireless Communications Mohammed Usman https://ebookbell.com/product/enabling-technologies-for-next- generation-wireless-communications-mohammed-usman-22540410
- 3. Enabling Technologies For Wireless Ebusiness 1st Edition W Kou https://ebookbell.com/product/enabling-technologies-for-wireless- ebusiness-1st-edition-w-kou-4268246 Enabling Technologies For High Spectralefficiency Coherent Optical Communication Networks 1st Edition Xiang Zhou https://ebookbell.com/product/enabling-technologies-for-high- spectralefficiency-coherent-optical-communication-networks-1st- edition-xiang-zhou-5677142 Enabling Technologies For Mobile Services The Mobilife Book Mika Klemettinen https://ebookbell.com/product/enabling-technologies-for-mobile- services-the-mobilife-book-mika-klemettinen-978324 Enabling Technologies For The Successful Deployment Of Industry 40 1st Edition Antonio Sartal Editor Diego Carou Editor J Paulo Davim Editor https://ebookbell.com/product/enabling-technologies-for-the- successful-deployment-of-industry-40-1st-edition-antonio-sartal- editor-diego-carou-editor-j-paulo-davim-editor-11909990 Enabling Technologies For The Internet Of Things Wireless Circuits Systems And Networks Sergio Saponara https://ebookbell.com/product/enabling-technologies-for-the-internet- of-things-wireless-circuits-systems-and-networks-sergio- saponara-51781770
- 6. IET COMPUTING SERIES 65 Enabling Technologies for Smart Fog Computing
- 7. Other volumes in this series: Volume 1 Knowledge Discovery and Data Mining M.A. Bramer (Editor) Volume 3 Troubled IT Projects: Prevention and turnaround J.M. Smith Volume 4 UML for Systems Engineering: Watching the wheels, 2nd Edition J. Holt Volume 5 Intelligent Distributed Video Surveillance Systems S.A. Velastin and P. Remagnino (Editors) Volume 6 Trusted Computing C. Mitchell (Editor) Volume 7 SysML for Systems Engineering J. Holt and S. Perry Volume 8 Modelling Enterprise Architectures J. Holt and S. Perry Volume 9 Model-Based Requirements Engineering J. Holt, S. Perry and M. Bownsword Volume 13 Trusted Platform Modules: Why, when and how to use them A. Segall Volume 14 Foundations for Model-based Systems Engineering: From patterns to models J. Holt, S. Perry and M. Bownsword Volume 15 Big Data and Software Defined Networks J. Taheri (Editor) Volume 18 Modeling and Simulation of Complex Communication M.A. Niazi (Editor) Volume 20 SysML for Systems Engineering: A model-based approach, 3rd Edition J. Holt and S. Perry Volume 22 Virtual Reality and Light Field Immersive Video Technologies for Real- World Applications G. Lafruit and M. Tehrani Volume 23 Data as Infrastructure for Smart Cities L. Suzuki and A. Finkelstein Volume 24 Ultrascale Computing Systems J. Carretero, E. Jeannot and A. Zomaya Volume 25 Big Data-Enabled Internet of Things M. Khan, S. Khan and A. Zomaya (Editors) Volume 26 Handbook of Mathematical Models for Languages and Computation A. Meduna, P. Horáček and M. Tomko Volume 29 Blockchains for Network Security: Principles, technologies and applications H. Huang, L. Wang, Y. Wu and K.R. Choo (Editors) Volume 30 Trustworthy Autonomic Computing T. Eza Volume 32 Network Classification for Traffic Management: Anomaly detection, feature selection, clustering and classification Z. Tari, A. Fahad, A. Almalawi and X. Yi Volume 33 Edge Computing: Models, technologies and applications J. Taheri and S. Deng (Editors) Volume 34 AI for Emerging Verticals: Human–robot computing, sensing and networking M.Z. Shakir and N. Ramzan (Editors) Volume 35 Big Data Recommender Systems Vols 1 and 2 O. Khalid, S.U. Khan and A.Y. Zomaya (Editors) Volume 37 Handbook of Big Data Analytics Vols 1 and 2 V. Ravi and A.K. Cherukuri (Editors) Volume 39 ReRAM-based Machine Learning H. Yu, L. Ni and S.M.P. Dinakarrao Volume 40 E-learning Methodologies: Fundamentals, technologies and applications M. Goyal, R. Krishnamurthi and D. Yadav (Editors) Volume 44 Streaming Analytics: Concepts, architectures, platforms, use cases and applications P. Raj, C. Surianarayanan, K. Seerangan and G. Ghinea (Editors) Volume 44 Streaming Analytics: Concepts, architectures, platforms, use cases and applications P. Raj, A. Kumar, V. Garcı́a Dı́az and N. Muthuraman (Editors) Volume 46 Graphical Programming Using LabVIEWTM : Fundamentals and advanced techniques J.C. Rodrı́guez-Quiñonez and O. Real-Moreno Volume 53 Nature-inspired Optimization Algorithms and Soft Computing: Methods, technology and applications for IoTs, smart cities, healthcare and industrial automation R. Arya, S. Singh, M.P. Singh, B.R. Iyer and V.N. Gudivada (Editors) Volume 54 Intelligent Network Design Driven by Big Data Analytics, IoT, AI and Cloud Computing S. Kumar, G. Mapp and K. Cergiz (Editors) Volume 56 Earth Observation Data Analytics Using Machine and Deep Learning: Modern tools, applications and challenges S. Garg, S. Jain, N. Dube and N. Varghese (Editors)
- 8. Volume 57 AIoT Technologies and Applications for Smart Environments M. Alazab, M. Gupta and S. Ahmed (Editors) Volume 60 Intelligent Multimedia Technologies for Financial Risk Management: Trends, tools and applications K. Sood, S. Grima, B. Rawal, B. Balusamy, E. Özen and G.G.G. Gan (Editors) Volume 61 Access Control and Security Monitoring of Multimedia Information Processing and Transmission Z. Lv, J. Lloret and H.H. Song (Editors) Volume 63 Personal Knowledge Graphs (PKGs): Methodology, tools and applications S. Tiwari, F. Scharffe, F. Ortiz-Rodrı́guez and M. Gaur (Editors) Volume 64 Intelligent Multimedia Processing and Computer Vision: Techniques and applications Shyam Singh Rajput, Chen Chen and Karm Veer Arya (Editors) Volume 115 Ground Penetrating Radar: Improving sensing and imaging through numerical modeling X.L. Travassos, M.F. Pantoja and N. Ida
- 9. This page intentionally left blank
- 10. Enabling Technologies for Smart Fog Computing Kuldeep Singh Kaswan, Jagjit Singh Dhatterwal, Vivek Jaglan, Balamurugan Balusamy and Kiran Sood The Institution of Engineering and Technology
- 11. Published by The Institution of Engineering and Technology, London, United Kingdom The Institution of Engineering and Technology is registered as a Charity in England & Wales (no. 211014) and Scotland (no. SC038698). † The Institution of Engineering and Technology 2024 First published 2023 This publication is copyright under the Berne Convention and the Universal Copyright Convention. All rights reserved. Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may be reproduced, stored or transmitted, in any form or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publisher at the undermentioned address: The Institution of Engineering and Technology Futures Place Kings Way, Stevenage Hertfordshire SG1 2UA, United Kingdom www.theiet.org While the authors and publisher believe that the information and guidance given in this work are correct, all parties must rely upon their own skill and judgement when making use of them. Neither the authors nor publisher assumes any liability to anyone for any loss or damage caused by any error or omission in the work, whether such an error or omission is the result of negligence or any other cause. Any and all such liability is disclaimed. The moral rights of the authors to be identified as authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988. British Library Cataloguing in Publication Data A catalogue record for this product is available from the British Library ISBN 978-1-83953-749-3 (hardback) ISBN 978-1-83953-750-9 (PDF) Typeset in India by MPS Limited Printed in the UK by CPI Group (UK), Eastbourne Cover Image: Olena T./E+ Collection via Getty Images
- 12. Contents About the authors xvii 1 Introduction of fog computing 1 1.1 Introduction 1 1.1.1 History of fog computing 1 1.1.2 Concept of fog computing 2 1.1.3 What is fog computing 3 1.1.4 Why is fog computing 3 1.2 How fog computing works 4 1.3 Taxonomy of fog computing 5 1.4 Fog computing versus cloud computing 14 1.5 Fog computing and IoT 15 1.5.1 Are fog computing and edge computing the same? 17 1.6 Fog deployment model 18 1.7 Distributed with the fog 20 1.8 Fog service models 21 1.8.1 Characteristics of the workload 25 1.9 Merits of fog computing 26 1.10 Demerits of fog computing 26 1.11 Application of fog computing 27 1.12 Conclusion 27 References 27 2 Fog computing in the IoT environment 31 2.1 Introduction 32 2.2 Models of fog computing 33 2.2.1 Basic structure of fog computing 34 2.2.2 Analyzing the literature with the help of scientific publications 36 2.3 Protocol of fog computing 38 2.3.1 Generic fog-computing architecture 39 2.3.2 Fog-computing environment model 47 2.3.3 Advanced fog-computing architecture 49 2.3.4 Fog-computing tree model 50 2.4 Conclusion 52 References 52
- 13. 3 Enhance quality of fog-computing environment using SDN and NFV technology 55 3.1 Introduction 55 3.2 The paradigm of fog/edge computing 56 3.3 What are fog nodes? 60 3.4 Connectivity technologies 63 3.5 Structure and behavior of fog computing 66 3.6 Analytic and performance parameters for fog/edge nodes 68 3.7 Performance enhancement techniques 69 3.8 Fog computing use cases 74 3.9 Key characteristics of fog computing 76 3.10 Challenges of fog environments 77 3.11 5G connecting technology and the fog architecture 81 3.12 Network Function Virtualization (NFV) and Software-Defined Networking (SDN) 83 3.13 Fog data analytics and use cases 86 3.14 Advanced fog-computing applications 87 3.15 Conclusion 91 References 91 4 Using P2P pervasive grid improves tunneling architecture and routing scalability 93 4.1 Introduction 93 4.2 Traditional and pervasive environments 94 4.2.1 Traditional fog-computing environment 94 4.2.2 Pervasive environment 95 4.2.3 Background 96 4.2.4 Pervasive grid in IIoT fog computing 96 4.2.5 Improved tunneling architecture 96 4.2.6 Enhanced routing scalability 96 4.2.7 Case study analysis 96 4.3 Proximity services with edge computing and pervasive grids 97 4.3.1 Ubiquitous networks 97 4.3.2 Cutting-edge computing technologies 97 4.3.3 Advantages of connected data processing 98 4.4 Developing a platform for edge and pervasive computing 98 4.5 Coordination and clustering 99 4.6 Data access 101 4.7 Context and scheduling 102 4.8 Monitoring ozone events for UV alerts 103 4.9 Input preprocessing 104 4.10 Time-series analysis, OSE detection and forecast 105 4.11 The APT protocol 106 4.12 Design principles 108 4.13 How APT works 109 viii Enabling technologies for smart fog computing
- 14. 4.14 Default mappers 110 4.15 Mapping information 111 4.16 Data forwarding 111 4.17 Failure detection and recovery 112 4.18 Mapping distribution protocol 113 4.19 Cryptographic protection 114 4.20 Incremental deployment 115 4.21 Routing policy and mapping 116 4.22 Conclusion 118 References 118 5 Vehicular fog computing and virtualization 121 5.1 Introduction 122 5.2 Vehicular and virtualization fog computing 123 5.2.1 Virtualization in fog computing 123 5.2.2 Vehicular nodes 124 5.2.3 Roadside infrastructure 125 5.2.4 Fog nodes 125 5.2.5 Virtualization layer 125 5.2.6 Data processing and analytics 125 5.2.7 Communication infrastructure 125 5.2.8 Applications and services 125 5.3 Vehicular mobility models 126 5.3.1 Selection of mobility model 126 5.3.2 Data collection and preprocessing 126 5.3.3 Model initialization 126 5.3.4 Simulation execution 126 5.3.5 Data analysis and evaluation 127 5.3.6 Iterative refinement 127 5.4 Content delivery and caching 127 5.4.1 Content caching in vehicular fog computing 127 5.4.2 Cache placement and management 127 5.4.3 Content delivery in vehicular fog computing 128 5.4.4 Virtualization for content caching and delivery 128 5.4.5 Performance evaluation and analysis 128 5.5 Network function virtualization 128 5.5.1 Network function virtualization (NFV) overview 128 5.5.2 Vehicular NFV use cases 129 5.5.3 NFV infrastructure in vehicular fog computing 129 5.5.4 NFV orchestration and management 129 5.5.5 Performance evaluation and analysis 129 5.6 Software-defined networking 130 5.6.1 Overview of software-defined networking (SDN) 130 5.6.2 SDN use cases in vehicular fog computing 130 5.6.3 SDN controller and network operating system 131 Contents ix
- 15. 5.6.4 OpenFlow protocol and southbound interfaces 132 5.6.5 Northbound interfaces and application ecosystem 132 5.6.6 Performance evaluation and analysis 132 5.7 Merits of vehicular and virtualization fog computing 132 5.7.1 Vehicular fog computing—low latency and real-time responsiveness 132 5.7.2 Vehicular fog computing—improved scalability and resource utilization 132 5.7.3 Vehicular fog computing—enhanced reliability and resilience 133 5.7.4 Virtualization in fog computing—efficient resource management 133 5.7.5 Virtualization in fog computing—service agility and rapid deployment 133 5.7.6 Virtualization in fog computing—improved fault isolation and security 133 5.8 Application of vehicular and virtualization fog computing 134 5.8.1 Dedicated short-range communication 134 5.8.2 Cellular-V2X 134 5.8.3 Internet of Things (IoT) protocols (e.g., MQTT and CoAP) 134 5.8.4 Network virtualization protocols (e.g., OpenFlow) 135 5.9 Quality of service 135 5.9.1 Vehicular fog computing and QoS 135 5.9.2 Virtualization in fog computing and QoS 135 5.10 Research vehicular and virtualization in fog computing 136 5.10.1 Resource management and optimization 136 5.10.2 QoS and service provisioning 136 5.10.3 Security and privacy 136 5.10.4 Vehicular mobility modeling and analysis 137 5.10.5 Edge intelligence and machine learning 137 5.10.6 Standardization and interoperability 137 5.10.7 Energy efficiency and sustainability 137 5.11 Conclusion 137 References 138 6 Smart XSS attack surveillance system for fog computing 141 6.1 Introduction 142 6.2 Cross-site scripting (XSS) attack 143 6.3 Key contribution XSS 145 6.3.1 Related work 146 6.4 Smart XSS attack monitoring system 147 6.4.1 Extracted web page module 148 6.5 Smart XSS surveillance system 149 6.5.1 Different types of intelligent XSS monitoring systems 150 6.5.2 Cloud data centers’ learning mode 151 x Enabling technologies for smart fog computing
- 16. 6.5.3 Online mode of virtualized network of fog computing 153 6.6 Modes of smart XSS monitoring system 154 6.6.1 Analyzer of the input field 156 6.6.2 Code embedded in JavaScript 157 6.6.3 Analysis on JavaScript 159 6.6.4 Protocol generation 159 6.6.5 Comment Comparison Tool for JavaScript 160 6.6.6 Context Finder 162 6.7 Components of smart XSS monitoring system 163 6.7.1 Conceptualization, development, and testing 164 6.8 Performance analysis 165 6.9 Conclusion 166 References 167 7 Spectral image analysis with supervised feature extraction 169 7.1 Introduction 170 7.1.1 Supervised feature extraction 170 7.1.2 Evaluation of supervised feature extraction methods 171 7.2 Remote sensing imaging 171 7.2.1 Types of remote sensing images 171 7.2.2 Hyperspectral remote sensing images 171 7.2.3 Supervised features extraction 172 7.3 Hyperspectral imaging and dimensionality reduction 173 7.3.1 Math functions 174 7.4 Supervised feature extraction in hyperspectral images 174 7.4.1 Modified Fisher’s linear discriminant analysis (MFLDA)-based feature 175 7.4.2 Prototype space feature extraction (PSFE) method 176 7.4.3 Maximum margin criterion (MMC)-based feature extraction method 177 7.4.4 Partitioned maximum margin criterion-based supervised feature extraction method 178 7.4.5 Hyperspectral feature partitioning 179 7.5 Experimental evaluation 183 7.5.1 Description of datasets 183 7.5.2 Performance measures 185 7.5.3 Parameter measures modified MMC-based approach 186 7.6 Conclusion 187 References 187 8 Developing of fog computing using sensor data 189 8.1 Introduction 189 8.2 Analysis of sensor data formats in smart cities 190 8.2.1 Sensor data in the SmartME Project 192 Contents xi
- 17. 8.2.2 Sensor data in the CityPulse project 211 8.2.3 Sensor data in the smart city 213 8.3 Pre-cleaned datasets for exploration in the Internet of Things, fog, and cloud 215 8.4 Conclusion 216 References 216 9 Information sharing on mobile IoT-based content aware smart home with fog computing 219 9.1 Introduction 220 9.2 Computation offloading 221 9.3 Result routing 222 9.4 Load balancing and efficient deployment 223 9.5 Mobility-aware edge computing 224 9.6 System design 225 9.6.1 Selection of potential employees 226 9.6.2 Networking with coworkers 227 9.7 Context-aware work stealing scheme 229 9.7.1 Extension for context-awareness 229 9.7.2 Task management skills that are both reactive and proactive 230 9.7.3 Performance of Docker image transfer 231 9.7.4 Transfer of Docker images 232 9.7.5 Distribution of tasks to fog nodes 233 9.7.6 Distribution of responsibilities 234 9.8 IoT-based smart home 235 9.9 Smart home scenario 236 9.10 ICON is an IoT-based, layered architecture 237 9.10.1 ICON’s design principles 239 9.11 Conditional logic with predicates 240 9.12 Implementation of ICON 242 9.12.1 The fog-computing system architecture for the ICON-based smart house 243 9.13 Conclusion 244 References 245 10 Security and privacy challenges in fog computing 247 10.1 Introduction 247 10.2 Fog application management 248 10.2.1 Application performance 249 10.2.2 Approach distributed data flow 250 10.3 Fog Big Data base analysis 251 10.3.1 Processing of streaming data 252 10.3.2 Big Data, Stream Data Analysis, and fog computing 253 10.3.3 Big Data, Stream Data, and the fog ecosystem: machine learning 253 xii Enabling technologies for smart fog computing
- 18. 10.3.4 Learning under guidance 254 10.3.5 Decision trees with distributed nodes 255 10.3.6 Methods for clustering large data 255 10.3.7 Tools like DBSCAN and DENCLUE are developed for use in Big Data environments 256 10.3.8 Tree-based incremental clustering 256 10.3.9 Mining association rules in large datasets with a P2P distributed computing architecture 257 10.3.10 Associative mining in real time 257 10.3.11 Methods for extensive learning 258 10.3.12 Large-scale datasets and advanced machine learning 261 10.3.13 Scale-up models 265 10.3.14 Different approaches to fog analytics 265 10.3.15 Other goods and services 266 10.3.16 ParStream 267 10.3.17 Cloud-based analytics in the periphery 267 10.4 Cloud Security Ontology 268 10.4.1 Create an ontology for safer cloud computing 270 10.4.2 Ontology: what it is and why it matters 271 10.4.3 CSO architecture as it is defined and operationalized 272 10.4.4 Cloud computing security requirements 273 10.4.5 Non-repudiation 274 10.4.6 Conceptual software architecture 274 10.4.7 Domain and scope determination for CSO 275 10.4.8 Identify the ontology’s imperative keywords 275 10.5 Fog security and privacy 276 10.6 Conclusion 279 References 280 11 Fog robotics 281 11.1 Introduction 282 11.2 Fog robotics 284 11.2.1 Facilitating distributed and shared learning 286 11.2.2 Data security, confidentiality, and ownership 287 11.2.3 Adaptability in resource allocation and placement 289 11.3 Comparison of fog and cloud robotics 290 11.3.1 The fundamentals of FR design 292 11.3.2 D2D communication in the FR architecture 293 11.3.3 In an FR architecture with many fog robot servers 294 11.3.4 Delivery of social robots in this scenario 295 11.4 Deep learning-based robotics 296 11.4.1 Transferring simulation learning to the real world 297 11.4.2 Execution environment on a networked system 298 11.5 Fog robot architecture 299 11.6 Implementation of fog robotics 301 Contents xiii
- 19. 11.7 Networking system with execution environment 301 11.8 Advanced robotics using fog computing 302 11.9 Applications of fog robotics 303 11.10 Conclusion 304 References 304 12 Cybernetic intelligence in fog computing 307 12.1 Introduction 308 12.2 A model of cybernetic intelligence in fog environment 309 12.3 Data utility in cyborg 310 12.3.1 Intelligent distributed computer network 311 12.3.2 A model for data-intensive applications in fog environment 312 12.3.3 Resources 314 12.3.4 Data source 316 12.3.5 Tasks 318 12.3.6 Use of data in cloud computing 319 12.3.7 Situational aspects 320 12.3.8 Data utility 323 12.3.9 Data life cycle 324 12.3.10 Using the data utility model 325 12.4 Content-aware intelligent systems fog computing in cybernetics intelligence 329 12.4.1 Social media analytics 331 12.4.2 Technical intelligence 333 12.4.3 Measurement and signature intelligence 333 12.4.4 Human intelligence 334 12.4.5 Finding humming in the dark 335 12.5 Using fog/edge computing for context-aware intelligence 336 12.6 Types of cybernetics intelligence 337 12.7 Conclusion 338 References 339 13 Further application of fog computing 341 13.1 Introduction 342 13.2 Geospatial technology with fog computing and IoT in agriculture 344 13.2.1 System to monitor irrigation 345 13.2.2 Treatment for insect and disease problems 346 13.2.3 Controlled fertilizer usage 346 13.2.4 Monitoring of greenhouse gases 347 13.2.5 Cattle tracking and monitoring 348 13.2.6 Assert the need of tracking and farming systems monitoring 348 13.2.7 Agriculture and information and communications technologies 349 xiv Enabling technologies for smart fog computing
- 20. 13.2.8 IoT’s functions 349 13.2.9 Big data’s place in the Internet of Things 350 13.2.10 The Internet of Things and cloud and fog computing 351 13.2.11 Sensors associated with plants 351 13.2.12 The GPS’s function 352 13.3 Big Data-based intelligent fog computing 352 13.3.1 Computerized information management 353 13.3.2 Big Data analysis and processing 354 13.3.3 The role of the cloud in Big Data 354 13.3.4 Cloud computing and geospatial data 355 13.3.5 Big geographical data 356 13.3.6 Technical measures for thermostatic regulation 357 13.3.7 Efficient and eco-friendly structures are called “green buildings” 358 13.3.8 Geospatial data influences thermal comfort in buildings 358 13.3.9 Thermostatic consistency of building equipment 359 13.3.10 Building materials made locally 359 13.3.11 Technologies and protocols for the Internet of Things overview 360 13.3.12 Technologies of information and communication and of physical location 360 13.3.13 Detection and monitoring techniques 361 13.3.14 The Internet of Things in the cloud 362 13.3.15 IoT with the advent of Big Data and the cloud 362 13.3.16 Computing on the cloud: from fog to cloud 363 13.4 Fog computing in health-care systems 364 13.4.1 The proposed system’s core functionality discerning the equipment 365 13.4.2 Disposables identification, location, and tracking 365 13.4.3 Functionality and design of the proposed system 366 13.4.4 The future of networking devices 366 13.4.5 The field of network analysis 367 13.4.6 Implied hardware 368 13.5 Protecting individuality within the paradigm of a recommender system 368 13.5.1 The cloud-based recommendation service reference model 369 13.5.2 Definition of risk 370 13.5.3 Formulation of the issue 370 13.5.4 Modifications of the FMCP protocols 371 13.5.5 PRR protocol for private relevancy scoring 371 13.5.6 PGD protocol for private group discovery (private group discovery protocol) 372 13.6 Conclusion 372 References 373 Contents xv
- 21. 14 Future research directions 375 14.1 Introduction 376 14.2 Cloud of things: cloud–IoT integration 377 14.3 Conclusion 382 Index 383 xvi Enabling technologies for smart fog computing
- 22. About the authors Kuldeep Singh Kaswan is presently working in the School of Computing Science and Engineering, Galgotias University, Uttar Pradesh, India. His contributions focus on BCI, cyborg, and data science. His academic degrees and 13 years of experience working with global Universities, such as, Amity University, Noida, India, Gautam Buddha University, Greater Noida, India and PDM University, Bahadurgarh, India, have made him more receptive and prominent in his domain. He received his doctorate in computer science from Banasthali Vidyapith, Rajasthan. He has also received a D.Eng. from Dana Brain Health Institute, Iran. He has supervised three PhD graduates and presently supervising four PhD stu- dents. He is also a member of IEEE, Computer Science Teachers Association, New York, USA, International Association of Engineers, Hong Kong, professional member of Association of Computing Machinery, USA. He has number of pub- lications also in international and national journals and conferences. He is an editor, author, and review editor of journals and books with IEEE, Wiley, Springer, IGI, and River. Jagjit Singh Dhatterwal is presently working as an associate professor with the Department of Artificial Intelligence and Data Science Koneru Lakshmaiah Education Foundation, Vaddeswaram, Andhra Pradesh, India. He has also worked with Maharshi Dayanand University, Rohtak and PDM University, Bahadurgarh, Haryana, India. He has supervised many UG and PG projects for engineering stu- dents and is presently supervising one PhD student. He is also a member of the Computer Science Teachers Association (CSTA), New York, USA, International Association of Engineers, Hong Kong, IACSIT, a professional member of the Association of Computing Machinery, USA, IEEE, and a life member of the Computer Society of India. His areas of interest include artificial intelligence, BCI, cyborgs, and multi-agents technology. He has a number of publications in inter- national and national journals and conferences. Vivek Jaglan is working as a professor and director at Amity School of Engineering and Technology, Amity University, Gwalior, India. He has nearly 19 years of teaching and research experience. His current research areas cover artificial intelligence, neural networks, fuzzy logic, and IoTs. He has presented and published 80+ papers in journals and conferences. He holds a doctorate degree from the Computer Science and Engineering Department, SGV University, Jaipur, India. He has supervised 7 PhD students, 11 masters’ students completion and is currently
- 23. supervising four PhD students. He has two design patents (India), one of the patents “Tooth Brush With Digital Display” design patent, which includes unique features that allow the brush to dispense only the required amount of toothpaste recom- mended by the dentist and verify that the user has properly brushed their teeth. He has been invited as an expert in the field of artificial intelligence and its approach on multiple occasions. Balamurugan Balusamy is a professor at the School of Computing Science and Engineering, Galgotias University, India. His research focuses on blockchain and IoT. He has published 30 technology books and over 150 journal and conference papers and book chapters. He serves on the advisory committee for several start-ups and forums and does consultancy work for the industry on Industrial IoT. He has given over 175 talks at events and symposiums. He is a member of several asso- ciations including IEEE and ACM. He holds a PhD degree on “Investigations of cloud computing access control techniques” from VIT University, Vellore, India. Kiran Sood is a professor at Chitkara Business School, Chitkara University, Punjab, India; an affiliate professor in the faculty of Economics Management and Accountancy at the University of Malta; and a postdoc researcher in the faculty of Applied Sciences at the University of Usak, Turkey. Her areas of research cover the fields of big data and finance. She serves as an editor for several refereed journals including the IJBST International Journal of BioSciences and Technology and the International Journal of Research Culture Society. She earned her doctor of phi- losophy in commerce with a concentration on product portfolio performance of general insurance companies from Panjabi University, Patiala, India. xviii Enabling technologies for smart fog computing
- 24. Chapter 1 Introduction of fog computing Abstract Purpose: In this chapter, including its history, introductions, benefits, dis- advantages, applications, and conclusion. Methodology: A decentralized computing facility known as pervasive computing, fog networking, or excessive moisture is one in which data, calculations, storage, and implementations are generated during the most suitable and appropriate place on-premises data centers and the internet. Findings: Other names for fog computing include accumulation of dust and fog networking. The notion of pervasive computing is essentially an extension of cloud-based solutions and the advantages it delivers to the network’s edges. Practical implication: This brings the benefits and capabilities of the cloud much closer to the location where data are produced and responded upon. 1.1 Introduction Cloud computing that is extended to the perimeter of an organization’s network is referred to as “fog computing,” which is a phrase that was developed by Cisco. In certain circles, in addition to the term “edge computing,” it is also known as “per- vasive computing.” The performance of computation, storage, and communications services may be facilitated more easily between network elements and the data cen- ters that handle cloud-based applications thanks to cloud environment. The term “fog computing,” which is synonymous with “fog networking,” refers to a decentralized computing environment in which computing, storage, and enterprise applications are made available in perhaps the most reasonable and realistic place at any moment along with the continuous spectrum from the data provider to the cloud in Figure 1.1. Fog computing is also sometimes referred to as edge computing [1]. 1.1.1 History of fog computing The concept of fog computing can be traced back to 2012 when Cisco first intro- duced the term. They envisioned fog computing as a paradigm that extends cloud computing to the edge of the network, enabling a seamless connection between cloud data centers and end devices. The goal was to address the growing demand
- 25. for low-latency, high-bandwidth applications in the era of the Internet of Things (IoT) and other data-intensive technologies. In the following years, fog computing gained traction as various tech compa- nies and researchers started exploring its potential. By 2014, the OpenFog Consortium was formed, a collaborative effort among academic institutions, industry leaders, and governmental organizations to standardize and promote fog- computing architectures. This further fueled the development of fog-computing technologies, leading to their integration into diverse fields, such as smart cities, healthcare, autonomous vehicles, and industrial automation. Fog computing has been a hot topic in the IT industry in recent years, and on November 19, 2015, the OpenFog Consortium was formed. Jeff Faders, Intel’s IoT Strategist, is the consortium’s first president and Cisco’s Sr. Managing Director Helder Antunes is its first chairperson [2]. As of 2021, fog computing continues to evolve and solidify its position as a vital component of the distributed computing landscape, enabling efficient and responsive data processing at the edge of the network. 1.1.2 Concept of fog computing ● A considerable volume of data is generated by IoT applications. These data have to be analyzed in order to make interoperability choices and to carry out a variety of activities. ● The cloud presents a variety of concerns, such as congestion, extremely high- bandwidth use, delays in real-time answers, and centralized data placement, when these data are sent. ● Cisco invented the phrase “fog computing” in 2012 in order to address these issues encountered by wireless sensor networks in the cloud infrastructure. Connected Server Sharing Information Storage Capacity Provider Cloud Data Figure 1.1 Fog computing 2 Enabling technologies for smart fog computing
- 26. ● To minimize latency and maximize spectral efficiency, it proposes to move processing closer to the end devices. ● The proliferation of sensor-based gadgets has resulted in an enormous volume of data. Nonvolatile memory and processing are both required for these data. Difficult, expensive, and time-consuming: preserving data in the cloud. It reduces operational time and costs by locating capabilities close to the equip- ment that will be used [3]. 1.1.3 What is fog computing The phrase “fog computing,” invented by Cisco, is often used interchangeably with “edge computing.” A dense computing framework at the edge of the network has been referred to as a fog foundation. Such systems are said to include features, such as reduced latency, reduced latency, and wireless connectivity. Real-time analytics and enhanced security are among the advantages. A fog-computing infrastructure, on the other hand, would be able to analyze everything from the network center to the edge of the network. Using fog computing, a system may adjust its signals depending on traffic monitoring in order to avoid accidents or minimize traffic congestion. Cloud-based analytics may also be used to store data for extended periods of time [4]. Cisco’s other examples involve rail safety, smart transmission and distribution restoration, and information security. In addition to key enabling technologies like interactive lighting and smart transportation meters, PrismTech Vortex cites vehicle-to-vehicle and vehicle-to-cloud connectivities. Cisco has provided an example of the analytics that may be conducted along a fog network in the picture below. 1.1.4 Why is fog computing IoT services, such as lower transmission assistance, situational awareness, and geo- distribution, are among the key goals of fog computing [2]. Many operations that need minimal delay between IoT devices and the closest fog web service or cloud service for local data analysis created may benefit from fog computing’s ability to extend cloud data center capabilities, such as computation, storage, and networking equipment [5]. The number of fog-based applications is increasing. Unconventional applica- tion framework requires considerable platform features that can only be provided if the program is compiled close to the end-users if new use cases for the fog envir- onment are to be realized. All the referenced applications are studied, and their justifications for employing a fog platform are identified. Augmented reality games, for example, need a latency of less than 10–20 ms for end-to-end latencies (connectivity and processor delay included). Propagation delay from 20 to 40 ms (over communication links) to up to 150 ms (over wireless networks), the distance between an end-user and the closest cloud data center (over 4G mobile networks). As a result, real-world usage of these apps is impossible. Run on the internet: Deploying the server element of these apps in fog plat- forms would be an easy way to lower the overall latency [6]. Video security cam- eras and other edge devices generate enormous amounts of raw data on a daily Introduction of fog computing 3
- 27. basis, which necessitates optimization of their available bandwidth. Huge network traffic is generated when so much data are sent to the cloud. Fog computing is an intermediary in reducing network traffic for these kinds of applications. In order to preprocess original information at the sources before transferring it to the cloud, fog middlewares are used. Smartphones, tablets, and other infrastructure components, such as smart IoT devices, have a limited amount of computational capacity. High-complexity applications like image retrieval take a long time to execute on these smart- phones. Delegation of authority for certain work to the intermediate virtual machines may affect productivity. When a cloud server is overloaded, the server- side of the running apps may be offloaded to fog servers. Alternatively, offloading might be another option. For many applications, privacy and security are of utmost importance. There is a large quantity of patient data that may be retrieved via E-health apps in healthcare administration. To ensure long-term accessibility, most recorded data are stored in the public cloud. However, many hospitals are concerned about the theft of per- sonal medical records. By offering storage space to the user or the hospital, private fog minimizes the problem of data privacy and security. Many IoT sensors and actuators may need sophisticated computers to run and control, such as implementation and condition monitoring, system implementation and switch on/off, and service distribution and incident management. By acting as a middleman, fog computing may supply computational power that not only makes it possible to operate various devices but also lets users tailor the services they get to their own needs [7]. Edge equipment, such as infusion pumps, heartbeat detection systems, and other monitoring devices, have improved in hospitals. As a result, the US suffers from the third-leading cause of mortality each year [8] due to difficulties inte- grating these devices with patients. Remotely hosted apps that interface directly with the monitoring devices allow for dynamic responses based on real-time data, thanks to fog computing. In the IoT context, one of the most pressing concerns is the amount of energy used by massive IoT devices. In order to save energy, fog computing allows these appliances to make intelligent choices, such as turning on/off/hibernating. A pay-as-you-go strategy, which is more common in conventional cloud computing, reduces costs by charging a flat rate per unit of utilization. One-time costs for procuring private fog resources may be better than cloud costs for appli- cations [9]. Reducing network traffic and enhancing response times may be accomplished in part via the use of content parallelization and networking protocols. 1.2 How fog computing works While edge embedded, systems create and obtain information; they lack the com- putational and storage capacity to execute sophisticated processing and machine 4 Enabling technologies for smart fog computing
- 28. learning activities. Despite the fact that cloud servers have the capacity to perform these things, they are frequently too far away to analyze the data and reply in a reasonable timeframe. In addition, when maintaining the confidentiality of data subject to rules in various countries, having all endpoints connect to and transfer raw information to the server through the internet might have anonymity, con- fidentiality, and legal concerns [10]. Fog reduces the quantity of data transferred to the cloud by doing processing on a data hub in a connected home or on a smart router or entry point. Fog net- working enables short-term predictive modeling at the edge, while the cloud han- dles the source of energy, long-term business intelligence. Fog networking does not replace cloud technology. 1.3 Taxonomy of fog computing Fog computing has suggested classifications, and this is what it looks like. Categorization of edge computing presents a categorization of the current fog- computing efforts. Network architecture is highlighted by the categorization in the following ways. Configuration of fog nodes: At the edge of the network, the nodes with diverse architectures and settings may support fog-computing infrastructure. ● Node-to-node communication: An edge network’s methods for coordinating nodal cooperation among several fog nodes. ● Metric for resource/service provisioning: How to provide resources and ser- vices in a cost-effective manner under a wide range of conditions. ● Goals for the quality service. By introducing edge computing as a middle layer among cloud applications and end applications, the Service Level Objectives (SLOs) were achieved. A network system is appropriate for the situation. Fog computing is an extension of previous computing paradigms that have been implemented in various networking systems. Security is a major issue. Fog computing’s security considerations vary depending on the situa- tion. Numerous taxonomic groupings are covered by existing methods and solutions. According to fog-computing features, this taxonomy does not accurately reflect the relative effectiveness of all of the many recommended approaches. Different execution settings, networking topologies, application characteristics, resource architecture, and so on are all taken into consideration and addressed in the work that has been viewed in this study. Due to the complexity of fog computing, it is almost impossible to pinpoint the most optimal solution in terms of structure, service, and security. In the existing studies, fog nodes have been categorized into five different types: servers, networking devices, cloudlets, base stations, and automobiles. In- house web servers may be seen at bus terminals, shopping centers, roads, and even in public parks. In the same way that being light in weight is similar to because of virtualization and cloud computing, these fog servers are cloud servers. The fog server is one of fog computing’s most critical functional properties. Fog servers are Introduction of fog computing 5
- 29. referred to as microservices, micro data centers, nano servers, and so on in certain studies based on their physical size, while they are classified as cache servers, cal- culation servers, storage servers, and so on in other papers based on their functions. Fog computing may benefit from a server-based node design that increases com- puting and storage capability. However, it restricts the extent to which the execution environment may be used [11]. ● Devices for connecting to the internet: It is feasible that Fog-computing infrastructures might be built using devices like routers, switches, and set-top boxes, in addition to their typical networking duties. Several modern switches and routers include a variety of system resources, such as CPUs, extensible main and secondary memory, and programing platforms. In addition to stan- dard hardware and software components, certain specialized network interface cards, such as intelligent gateways and IoT hubs, have been represented as fog nodes in other publications. Network devices deployed in a dispersed way increase the prevalence of pervasive computing, but the physical diversity of devices has a significant influence on the provisioning of services and resources. The cloudlet is a micro-cloud that resides in the middle of the end device, cloudlet, and cloud hierarchy. Cloudlets may be used to complement MCC by providing mobile device users with cloud-based services. According to a series of studies, cloudlets have been referred to as “Fog nodes.” A huge number of end devices may be handled concurrently using cloudlet-based fog computing. Cloudlets, although being deployed at the edge, may nonetheless function as centralized components in certain circumstances owing to struc- tural restrictions. There are still serious issues with fog computing that prevent it from supporting IoT [12]. ● Stations at the ground zero: An essential part of any wireless network, base stations process and transmit data to and from the mobile nodes. Traditional base stations equipped with particular storage and computation capabilities have recently been deemed viable for fog computing in recent research. Fog nodes may be created using RSUs, small cell access points, and so on, much as regular base stations. Fog-based extensions of cloud radio access network (CRAN), vehicular ad hoc network, and similar networks are better served by base stations. Fog creation using base stations, on the other hand, is compli- cated by high deployment costs and networking interference. ● Intelligent vehicles: Fog nodes may be placed in moving or stationary cars at the edge of a network with computing resources. It is possible to create a highly dispersed and scalable fog environment using vehicles. It will, however, be very difficult to provide privacy and fault tolerance while maintaining optimal quality of service (QoS) in such an environment. ● Collaboration at the node level: Cooperative approaches for cluster, peer-to- peer (P2P), and master/slave computation nodes for fog have all been reported in several studies. Cluster nodes in the network may form their own clusters in addition to maintaining a collaborative computing environment. When fog nodes are in close proximity to one another, they might form clusters. The creation of 6 Enabling technologies for smart fog computing
- 30. functional subsystems and congestion control may be prioritized while the nodes are forming a cluster. It is possible to use the capabilities of several fog nodes concurrently by using cluster-based cooperation. It’s tough to scale static clus- ters at runtime because dynamic cluster generation is heavily reliant on the current demand and readily available fog nodes. In both circumstances, the networking overhead is critical to the overall outcome. P2P cooperation is highly popular in fog computing because of the distributed nature of the system. It is possible to do P2P cooperation in a hierarchical or flat manner. There are a variety of ways to classify P2P cooperation between fog nodes besides proxi- mity. For example, virtual computer instances are shared among nodes in a cloud computing environment to optimize resource utilization and deliver scal- able and cost-effective solutions to users. It is about P2P cooperation rather than relying on a single node’s processed output alone. In P2P cooperation, fog nodes may be easily augmented and made reusable. P2P nodal cooperation, on the other hand, is plagued by problems with dependability and access control [13]. ● Master–slave: Master–slave nodal cooperation has been extensively discussed in a number of publications. A master fog node typically manages slave nodes’ functioning, processing load, environmental protection, information flows, and other aspects. The fog-computing environment may also create a hybrid col- laborative network using a master–slave strategy, cluster nodal connections, and P2P interactions. As a result, both the master and the slave fog nodes in order to process data in real time require high-bandwidth communication. There are several aspects that play a role in the supply of resources and ser- vices in fog computing, time, energy, application and database context, and more are all taken into account. ● Time: Fog computing makes efficient use of time as a key consideration in resource and service supply. The amount of time it takes to complete a job is known as computation time. It’s important to keep in mind that the amount of time it takes to operate an application relies heavily on the configuration of the resources it’s operating on. In addition, the time it takes to compute a task helps to distinguish between the current and previous periods of different programs and has a substantial impact on fog’s resource and power management. The time it takes for data items to be exchanged in a fog-computing environment is referred to as communication time. It has been explored in two ways in the literature: There is a direct connection between the end devices/sensors and fog nodes. To aid in task execution, the network context is reflected in the required commu- nication time. A system’s deadline sets the maximum amount of time it may go without receiving a service. Task completion satisfaction has been regarded as a significant QoS indicator in various studies. The delivering services constraint distinguishes between latency sensitivity and latency tolerance applications and games. Service access times in a multitenant cloud architecture, service reaction times, and other time-based metrics such as these may be studied for efficient function and resource deployment and management in sensor networks [14]. ● Data: Fog-computing literature makes frequent use of input data size and data flow characteristics, two data-centric metrics. The quantity of data that must be Introduction of fog computing 7
- 31. processed via fog computing is referred to as data size. There have been sev- eral discussions about the computational space needs of requests in relation to data size. In addition, data gathered from a large number of dispersed sensors and devices may have the characteristics of Big Data. Provisioning resources and services based on data load might be a useful strategy in this situation. Data size also has a significant effect on determining whether a computing activity should be performed locally or remotely. The properties of data transmission are defined by the data flow. Event-driven or real-time data flow in the fog-computing environment may have a significant impact on resource and service delivery. In addition, abrupt changes in data flow might lead to dynamic node load balancing. Fog computing’s resource and service provi- sioning may also be analyzed in terms of heterogeneous data architecture, data semantic norms, and data integrity needs [15]. ● Cost: Fog resource and service delivery may be heavily influenced by con- siderations relating to cost, both from the standpoint of service providers and customers. It is easy to see how broadband use and associated expenses have a direct impact on connectivity costs in a fog-computing system. Some studies attribute connectivity prices to the upload of data from end devices/sensors and the exchange of data across nodes, while others attribute networking delays caused by broadband problems to network security costs. In a fog-computing environment, deployment costs are mostly tied to the costs of setting up the infrastructure. In several studies, efficient resource and service supply has been linked to cost-effective infrastructure implementation. The cost of deploying infrastructure may be broken down into two parts: the actual deployment of fog nodes in the network, and the creation of virtual computing instances inside those nodes. Fog nodes’ computational costs when executing applications or processing activities are referred to as “execution costs.” The use of execution costs in resource provisioning and invoicing is rare in fog computing despite its widespread usage in other computing paradigms. Task completion time and resource utilization costs have been used to compute the overall cost of these tasks. Fog-computing resource and service provisioning may take into account migration costs as well as the previously listed costs, as well as charges for security precautions, the most a customer is willing to pay for a product or service [16]. ● Consumption of energy and environmental impact: In a few studies, fog resources and services have been prioritized based on energy concerns. Fog- cloud interaction has been studied extensively for its energy consumption across all devices, as well as the trade-off between energy and delay at various stages of the process. Previously, the carbon emission rate per unit energy consumption of various nodes was taken into account for resource provisioning objectives in another study. Fog resources may be provisioned according to the energy constraints of end devices/sensors, such as residual battery life and the energy characteristics of communication media. Context refers to the condi- tions under which a certain thing is found. For resource and service delivery, user and application context has been studied in fog-based research papers. In 8 Enabling technologies for smart fog computing
- 32. the future, resources may be allocated to a user based on the user’s attributes (e.g., service use history and service relinquish likelihood). Service and resource provisioning may benefit from customer input, such as net promoter score and customer needs. Service provisioning in previous works has taken into account user density, mobility, and network state. The context of an application may be defined as the operational needs of several applications. Prerequisites for task performance (such as processor speed, storage, and storage), as well as network connectivity, and other operational needs might impact the supply of resources and services. The present workload of various apps has also been taken into account as an application context in other research. Fog-computing contexts may also be considered in terms of the execution environment, node character- istics, application design, and so on, and these contexts can play a significant role in providing resources and services. As a result, every piece of background material must be thoroughly examined [17]. ● Goals for the quality of service: Certain SLOs have been suggested to be achieved using a variety of application platforms, computational models, and optimizations of fog node architecture approach in current research. Almost all of the successfully achieved SLOs are management-oriented and deal with concerns such as latency and power consumption as well as cost, resource allocation, data storage, and other types of applications. ● Latency control: Fog computing’s latency control essentially prevents the eventual service delivery time from exceeding a predetermined threshold for acceptable latency. A service request’s maximum acceptable latency or an application’s QoS requirement may fall within this barrier. Some efforts have placed an emphasis on efficient nodal cooperation start to guarantee that com- pute activities carried out by collaborated nodes may be completed within the latency limitation set. It has also been shown that distributing computing tasks across clients and fog nodes may reduce service request computation and com- munication delay. In addition, a low-latency fog network design was presented in another paper to control latency. The primary goal of this project is to find a node in the fog network that delivers services with the least amount of delay. ● Controlling the costs: Operating expenses (Operating Expenses) may be evaluated in terms of fog-computing cost management (OPEX). Distribution of fog nodes and their networks is the primary cause of CAPEX in fog com- puting. Fog computing’s CAPEX may be kept to a minimum by strategically placing and using an optimal number of fog nodes. According to this concept, the total cost of fog computing is reduced by optimizing the positioning and number of sensor nodes in use. Fog nodes are also referred to as virtual machines launch vehicles and virtual machines in another study. The cost of running data processing processes on these virtual machines varies from pro- vider to provider, and the cost is not always the same. To reduce OPEX in fog computing, it is possible to take advantage of the cost diversity of fog nodes/ providers. According to this, the article proposes a technique to discover the best fog nodes to host Virtual Machine (VMs) in order to reduce OPEX in fog computing [18]. Introduction of fog computing 9
- 33. ● Management of the network: Fog computing’s network management com- prises, for example, the control of network congestion in the core, the support of software-defined networking/network function virtualization (SDN/NFV), and the guarantee of smooth connection. Network congestion is primarily caused by an increase in network overhead. Because IoT devices/sensors are widely dispersed over the edge, the cost on the core network may be greatly increased by concurrent interactions between end components and cloud data centers. As a result, network congestion will arise, lowering the system’s overall performance. In light of this, a layered fog node design has been sug- gested that allows for the processing of service requests at the node level. As a result, despite getting large numbers of service requests, clouds only receive compressed versions of such requests, which have less impact on the network. There is a lot of interest in virtualizing the traditional networking infra- structure. Virtualized networks are made possible in large part by SDN. SDN is an SDN approach that separates control and data planes from communication gear and puts it into software on different servers. Support for NFV is a crucial feature of SDN. To put it simply, NFV is an architectural idea that enables conventional networking tasks to be virtualized so that they may be done via software. SDN and NFV have a significant impact on cloud-based environ- ments because of their large variety of services. New network topologies for fog computing have been developed to allow SDN and NFV as a result of this; as a result of their physical variety, end devices are able to communicate seamlessly with other entities such as the cloud or fog or desktop computers or mobile devices. As a result, finding resources and keeping the network’s communication and computing capacities up to date are made simpler. It is a problem that has already been addressed in fog computing, with new archi- tectures for fog nodes like the IoT hub and fog networking like the vehicular fog computing (VFC). Another development in fog computing is a policy- driven framework for ensuring secure connectivity between devices [19]. ● Management of computations: Fog computing’s SLOs include a high priority on ensuring that computational resources are properly managed. It is possible to esti- mate computer resources, distribute workloads, coordinate computing resources, and more with fog computing. Resources may be assigned according to certain rules in fog computing so that suitable resources can be allocated, desired QoS can be attained, and an exact service fee can be enforced. According to the current litera- ture, resource estimation strategies are built based on user characteristics experi- enced QoE, features of service-accessing devices, and so on. Fog computing’s workload distribution should aim to optimize resource usage while minimizing computational idle time. More specifically, a balanced load is ensured on several components. A scheduling-based workload allocation strategy has been imple- mented in a fog-based research project in order to distribute the computational burden across fog nodes and client devices. As a result, both parties’ overhead costs are reduced, which raises QoE. In a separate study, a framework for balancing fog- cloud communication delays and power consumption was proposed. Because of their heterogeneity and resource limitations, coordination among various fog 10 Enabling technologies for smart fog computing
- 34. resources is absolutely essential. With fog computing, large-scale applications may be distributed over several fog nodes due to its decentralized nature. Without ade- quate coordination of fog resources, it would be difficult to achieve the required performance under these situations. In light of this, a paradigm for managing fog resources based on directed graph coordination has been developed. ● Management of application programs: Efficient programing platforms are critical for successful fog-computing application administration. In addition to the scalability and compute offloading capabilities, these features aid application administration. Development, compilation, and execution of programs are all made easier with the help of a programing platform, which includes all of the components listed above. Due to fog computing’s dynamic nature, ensuring effective resource management and real-time decision-making capabilities are paramount for its successful implementation in applications such as IoT and edge computing. It’s quite tough to develop software to handle large-scale applications. Mobile fog, a new development platform, has addressed this issue. Mobile fog’s reduced abstractions of programing paradigms make it feasible to build large-scale decentralized applications. As well as the coordination of resources during the implementation, an implementation framework for fog computing was also established in another document. To keep their QoS high even as the number of app users increases and unexpected occurrences occur, apps must be capable of adapting. Application scheduling and service access for users may both benefit from scaling strategies. Fog computing has recently presented an architecture for a QoS-aware self-adaptive scheduler to facilitate the scalable scheduling of data stream applications. Using this scheduler, pro- grams may be scaled up and down based on the number of users and the amount of resources available. It is also simpler to arrange programs in a dispersed form because of the scheduler’s self-adaptive capabilities. Fog computing has also offered an adaptive approach for users to choose their service access mode depending on the distance, location, and QoS needs of the service-accessing entities. It is possible to transmit computational activities from resource- constrained end devices to more resource-rich devices via offloading methods. In a mobile cloud environment, computational offloading is a typical occurrence. As part of fog computing’s compatibility upgrade support for mobile applica- tions’ computation dynamic provisioning in other communication networks has been stressed in various publications lately. Mobile apps’ distributed computing and the availability of resources have been examined in these studies [8]. ● Management of data: Fog computing can’t function properly if its informa- tion systems SLOs aren’t in place. Different research studies have looked at fog computing’s data management takes a multifaceted approach. Fog com- puting’s information management approach places a high value on computa- tional intelligence services and preparatory distribution of resources. As an alternative, low-bandwidth aggregation from scattered end devices/sensors may be explored in the interest of improved data administration. End devices/ sensors, on the other hand, have limited storage capacity. Storage enhancement for end-entity data storage and data may have a substantial influence on fog Introduction of fog computing 11
- 35. computing in this case. Data processing in fog computing has also been emphasized as a crucial part of storage growth for smart applications. ● Management of energy resources: Power management may be provided as a service to various fog computing is being used to connect systems. A cloud infrastructure for fog computing might enable home-based distributed systems to regulate power with customized user control, according to a report. Centralized cloud data centers’ power usage may be managed in certain cir- cumstances using cloud services. Data center power consumption is mainly dependent on the sort of applications being executed. By offering infra- structure to support a number of energy-hungry applications, fog computing may complement cloud data centers in this situation. Therefore, cloud data centers will be able to maintain proper power management as a result of reduced energy consumption. Furthermore, carbon footprint emissions can be lowered by carefully controlling the power used in fog computing. ● Usefulness of networking systems: The IoT relies heavily on fog computing. Fog computing’s usefulness in different mobility, content management, radio access, and vehicle networks are all examples of communication networks that have been emphasized in recent research publications. ● Internet of Things: Every device in the IoT can communicate with each other and exchange data. The IoT environment can be viewed from many different angles. Additionally, in various fog-based research studies, this contact has been categorized as either industrial or home-based execution environments. Furthermore, fog-computing systems and service models have also taken into account many kinds of IoT, including networks of wirelessly sensing devices and cyber–physical systems. ● Access networks for mobile devices/mobile phones: Fog computing’s applicability to mobile networks has also been studied in a number of studies. Fog computing’s interoperability with 5G mobile networking has been a major focus of these studies. Compared to current cellular networks, 5G allows for substantially faster connectivity, more signal capacity, and reduced latency in service delivery. Fog computing may be used in various mobile networks other than 5G, such as 3G and 4G. A different study looked at how workloads in fog cloud for mobile communication are allocated depending on trade-offs between power and delay. Individual devices communicate with other net- work entities via radio connections in the radio access network. CRAN, the cloud-assisted Radio Access Network (RAN), has already piqued the interest of many researchers. Fog computing–based radio access networks have also been investigated as a possible supplement to the capabilities of CRAN [20]. ● Power line communications via passive optical network: With the introduction of LRPON, backhaul services for homes, businesses, and wireless networks can now take advantage of a new, low-latency, high-bandwidth technology for long- distance communications. LRPONs facilitate network consolidation in addition to providing a wide coverage area. Fog computing has been integrated with LRPONs in order to optimize the network design in this article. Communication over the smart grid’s power lines (PLC) is common. Data and alternating current are sent 12 Enabling technologies for smart fog computing
- 36. concurrently in PLC utilizing electrical connections. Discussion of fog-computing PLCs in electrical power distribution has been extensive. ● Internet-based platform for the distribution of content: Distributed proxy servers supply material and provide good reliability and availability for the end-users via a content distribution network. Many fog-based research projects use fog nodes as they open up opportunities to make content dissemination easier. Users may access fog-based content services with little latency since fog nodes are scattered over the network’s edge. The dissemination of high- quality information will run more smoothly as a result. ● Network of automobiles: Data interchange and resource augmentation are made possible by vehicular networks, which allow the autonomous establish- ment of a wireless communication network among cars. Computational and networking resources are made available to cars as part of this network. Fog nodes are vehicles that reside at the edge of a network and are used to promote a fog computing–based vehicular network in several studies [21]. ● Fear for one’s safety: Fog computing relies on an underlying network between end devices and cloud data centers; it has a high level of security risk. However, security considerations in cloud computing are essential to safeguard sensitive data and protect against potential threats or breaches, making robust encryption, access controls, and continuous monitoring critical components of a secure cloud environment. In the literature, fog computing has been resear- ched in terms of information identification, confidentiality, secure data exchange, denial-of-service (DoS) attack, and so on. ● Authentication: In fog-based systems, user authentication plays a critical role in preventing infiltration. Unwanted access to fog services is very intolerable since they are utilized on a “pay-as-you-go” basis. In addition to user authentication, the secure fog-computing environment has seen device authentication, data transfer authentication, and instance authentication. End device/sensor data are processed via privacy fog computing. Sometimes, these statistics are discovered to have a strong correlation with users’ personal and professional circumstances. As a result, one of the most pressing issues in fog computing is the protection of user security. It has been noted that privacy is an issue with fog-based vehicular computing. ● Data encryption: Fog computing is a useful adjunct to cloud computing. Cloud computing is required in certain circumstances for data handled via fog computing. Fog nodes must encrypt this data since they frequently include important information. In light of this, the proposed fog node design includes a data encryption layer. ● DoS attack: Because fog nodes have a limited amount of resources, they are unable to accommodate a high number of simultaneous queries. Fog comput- ing’s performance may suffer greatly in this situation. DoS attacks might be crucial in causing such significant service outages in fog computing. Fog nodes may be kept busy for a longer length of time by concurrently issuing a large number of irrelevant service requests. Due to the lack of resources, helpful services are no longer accessible. Fog computing has been used to discuss and clarify this kind of DoS attack. Introduction of fog computing 13
- 37. ● Inquiry into the gap and possible future directions: Rather than relying on remote servers, fog computing makes advantage of cloud resources that are already nearby. Fog computing is critical for supporting widely dispersed end devices and sensors. As a result, fog computing has emerged as a major research area in both academia and industry in recent years. There is an overview of several reviewed publications on fog computing. Many important aspects of fog computing have been discussed already, but there are still some issues that need to be addressed if this field is to advance further. In this section, we will talk about some of the gaps in the existing literature and possible future research directions [22]. Provisioning of resources and services in light of current situation: Fog comput- ing’s resource and service provisioning may benefit from context awareness. Fog computing may acquire contextual information in a variety of ways, such as ● Location, time (peak, off-peak), and so on. ● The application’s perspective: latency sensitivities, application design, and so on. ● Mobility, social connections, activity, and so on are all examples of user context. What resources are accessible on the device? How much juice is left in the battery? The context of a network might include factors, such as bandwidth and traffic. There are still many undiscovered features of background information, ignor- ing the fact that several fog-based research studies have taken into account specific information while assessing how resources and service operations might be studied using fog-based research methods [23]. 1.4 Fog computing versus cloud computing Many people use the words fog computing and edge computing alternatively because they both entail moving intelligence and processors closer to the location where data are produced. However, the primary distinction between the two is the location of intellect and computing capacity. ● Cloud technology: The processing of data and applications in the cloud is time-consuming for huge datasets. Bandwidth issues result from the trans- mission of all data over cloud channels. Because of distant servers, slow response times and scalability issues arise. ● Cloud computing: Instead of displaying and operating from a central cloud, fog operates at the network’s periphery. Therefore, it uses less time. Less need for bandwidth, since all data are pooled at a single access point as opposed to being sent through cloud channels. It is conceivable for a fog-computing platform to circumvent reaction time and scalability difficulties by selling tiny machines known as edge servers in direct user view. 14 Enabling technologies for smart fog computing
- 38. 1.5 Fog computing and IoT Do not be astonished to learn that about 31 billion IoT devices are now in operation. It is no surprise that we generate 2.5 quintillion bytes of data every day. Clearly, we need advantages over conventional techniques of data management. This is when fog computing comes into play. When an application or device accumulates huge amounts of data, effective data warehousing becomes difficult, not to mention expensive and difficult [24]. Heavy data place a strain on bandwidth usage. It is costly to build Big Data centers to store and arrange this data. Fog computing collects and distributes storage, computation, and network connection services, decreases energy consumption, increasing the productivity and value of the data, and decreases space and temporal complexity. Consider two IoT examples: ● Sustainable urban: Data centers are not designed to accommodate the rising demand for smart city applications. As more individuals began to use IoT devices, more information would be transferred and accessible. Fog computing may assist such inadequate facilities and smart grids in delivering the true benefit of IoT application development. ● Facilities: The word “utilities” encompasses applications like hospitals, rail- roads, and enforcement agencies that need the most advanced data delivery infrastructure to enhance their functioning. Knowledge on energy consump- tion, fractures, and water leaks, for instance, may be utilized to update payment information, save lives, and enhance operations. How cloud technology increases the value of solutions for the IoT and end-users is getting more potent. A significant proportion of data is currently processed in the cloud in Figure 1.2. In addition, here are six advantages that fog computing may pro- vide to the IoT design and development phase. ● Maximum business flexibility: With the proper tools, fog apps may be developed and deployed as required. These programs allow the user to custo- mize the device’s behavior. ● Enhanced security: As a proxy for devices with limited resources, fog com- puting updates their software and security credentials. It installs fog nodes with Huge Business Agility Better Security System Low-latency Rate Network Bandwidth Efficiency No Interruption Services Refined User Experience Figure 1.2 Fog computing enhances solution Introduction of fog computing 15
- 39. the same policies, processes, and controls as the rest of the IT system. When a large number of nodes process data in a complex distributed system, it is simpler to monitor the security status of adjacent linked devices. ● Low delay: Have you noticed how rapidly Alexa responds to requests? This is due to the reduced latency provided by fog computing. Since the “fog” is physically closer to all users (and devices), it is able to deliver instantaneous replies. This technology is suitable for all time-sensitive tasks. ● Frequency performance of the network: Fog computing offers rapid and effective data processing dependent on application requirements, available computer resources, and network connectivity. Instead of transmitting infor- mation via a single channel, information is integrated in several places. This decreases the amount of data that must be sent to the cloud, thus conserving network capacity and cutting prices significantly. ● Continuous services: Fog computing may operate autonomously and provide continuous service, although when cloud network access is impaired. Additionally, because of several linked channels, connection loss is very impossible. ● Enhanced user experience: Edge nodes use low-power technologies includ- ing Zigbee, Bluetooth, and Z-Wave. Fog computing provides immediate communication between devices and end-users, regardless of network access, hence boosting the user experience. Although fog and cloud computing may seem identical at first glance, they are really distinct levels of industrial IoT solutions. Here are a few distinctions between the two technology solutions: ● Architectonics: Fog architecture is dispersed and consists of millions of tiny nodes situated as near as possible to client devices. The design is both hier- archical and flat, with several levels creating a network. Cloud architecture, on the other hand, is centralized. Large data centers are dispersed over the world, which puts them in close proximity to mobile terminals. ● Communication between devices: Fog technology acts as an intermediary between software-defined data centers, bringing it closer to end-users. Without the fog barrier, the cloud encounters the electronics, resulting in a lengthy process. ● Data processing: In fog computing, data collection and processing occur near the information source, which is essential for real-time management. Fog determines whether to transfer its capabilities to the cloud to process infor- mation from many data sources. In a cloud computing approach, the same occurs through distant data centers far from the educational resource. ● Computational skills: Compared to fog, cloud technology technologies are more advanced. ● Quantity of nodes: In contrast to the cloud, fog comprises millions of little nodes. Due to the instantaneous reactivity between the equipment and end- users, analysis fog provides short-term analysis. Nevertheless, the cloud is designed for long-term study because of its poor reactivity. Because cloud computing employs a variety of security mechanisms and protocols, the danger of cyber-attacks and data loss is significantly reduced. In addition, it has a distributed architecture. Without an internet connection, cloud technology is impossible. Since it is also centralized, cyber risks are more likely. 16 Enabling technologies for smart fog computing
- 40. 1.5.1 Are fog computing and edge computing the same? This is a challenging question. To keep things simple, fog and edge computing are virtually the same. Both methods employ computational performance to bring intelligence back to the lowest area network level of the network architecture. This inhibits the execution of computing activities in the cloud, saving time, resources, and money. In addition, both cloud technology and edge computing may help organizations lessen their dependency on cloud-based platforms for data gathering, which lowers latency difficulties and the time required to make data- driven choices [25]. Perhaps one of the most notable distinctions is the data processing. With a fog node, data are packaged in fog. Computing at the edge interprets information in the system or sensor without transferring it to another infrastructure. With effective collecting and analyzing data in real time, however, both technol- ogies save time and money when it comes to sustaining procedures. Imagine receiving statistics in near real time that are valuable for improving performance and enhancing uptime. Both fog computing and networking technologies make this feasible. ● Fog computing and IoT app development in action: Fog computing stands out as a reliable, dynamic, and cutting-edge technology in a wide range of fields. In this part, we will look at four real-time examples: On the fog plat- form, data transfer for video streaming applications is well organized. Because of the flexibility and scalability of fog networking and real-time data proces- sing, this is possible. In addition, fog encourages interaction in a virtual full- featured system, allowing real-time video business intelligence for security cameras [26]. ● Monitor and control systems in the healthcare sector: Future healthcare choices would be incomplete without comprehensive and real-time health data. Data transfer in real time is feasible; however, thanks to the implementation of fog-computing frameworks. “U-Fall” is another key use case, since it auto- matically identifies a large fall in the event of mild strokes. ● Playing video games: Fog computing, like the cloud, puts computational power closer to the players’ fingertips. It is no secret that SEGA’s fog gaming system relies on the low latency provided by local gaming arcades and centers. So, instead of streaming from the cloud, players would use the local arcade equipment’s CPUs to power up their games. The dispersed devices improve the quality of the online gaming experience for many players: a system for intel- ligent traffic light control; imagine a traffic signal system that is smart uses fog nodes. Multiple sensors on the node interact locally to detect the presence of bicyclists or walkers, as well as the speed and distance traveled by cars. The green light sends out warnings based on the information. Since it already monitors video security cameras, an ambulance may be easily seen by its emergency light and warning alarm. Allowing a vehicle to pass traffic may be done by adjusting the traffic signals in the area [27]. ● In a nutshell: The IoT solution generates enormous amounts of data every day, and fog computing serves as a partner to the cloud. Data processing near Introduction of fog computing 17
- 41. the source of information overcomes the difficulties of growing data volume, velocity, and diversity as previously stated. It provides companies with more control over their data. Additionally, fog computing enables a more rapid understanding of and reaction to occurrences. A cloud-based analysis is no longer required. This eliminates the need to offload large amounts of data to the core network, which saves money on network traffic. To secure sensitive IoT datasets, fog computing analyzes them inside a company’s firewall. In the end, this leads to better company agility, security, and service quality. 1.6 Fog deployment model Fog models may be classified depending on who owns the fog infrastructure and the resources underneath it. In an organization, a third party or a combination of them is responsible for creating, managing, and operating a private fog. It may be installed either on- or off-site. A single entity has exclusive access to the resources of a private fog (e.g., business units). A firm, academic institution, government agency, or a combination of these, creates, owns, manages, and operates a public fog. Fog service providers have it installed on their property. Public fog materials are available to the entire public for free. Several groups in the community, as well as a third party, or a mix of the three, are involved in the creation, management, and operation of the community fog. Typically, the materials are made available only to members of a certain community of organizations that share the same set of issues. It is possible to combine general populace fog computing with cloud applications in a hybrid fog-computing model (i.e., hybrid cloud). Since the fog is devoid of physical resources, it may be advantageous to use this method. The infrastructure has been relocated to a hybrid cloud in an attempt to enhance performance. A cloud infrastructure is expandable, elastic, and capabilities may be accessed on-demand if needed. Because of this, apps depend on the fog models used to deploy them. There are 13 private fog and 17 hybrid fog apps evaluated, with the majority being deployed in a private fog. It is also worth noting that none of these apps are cur- rently being used in public or communal fog. Astonishingly, the applications may be grouped depending on their specific needs and the capabilities of the fog models. It is for these reasons that we conclude that a private fog is the best option. As a result of the high risk to privacy and security posed by apps dealing with personal data, such as wearable devices, many of these applications are better suited to be deployed in a private fog cloud controlled by the user or a third party that the user has confidence in. Similarly, for reasons of security, many companies choose to use a secure cloud to operate autonomous robotic applications. For applications that are sensitive to latency and resource needs, such as web hosting, the fog platform is the ideal solution. In the classic cloud approach, you pay only for what you use, which is called a pay-as-you-go (PaaS) model. Some programs, especially those that do not need a lot of flexibility or administration, are more cost-effective to run on-premises. One-time costs for private fog capabilities are less expensive than typical cloud services for 18 Enabling technologies for smart fog computing
- 42. similar programs. Fog hybrids are designed to scale fog infrastructures’ resources, so they may be used for applications that otherwise would be unaffordable (i.e., com- putation and storage). As a result, access networks serve as essential building ele- ments for fog platforms since they link IoT devices to the platform. An ultra-low latency and massive data volume network architecture are needed to process and respond to data produced by applications in microseconds. Fog environments allow for the deployment of a wide range of standards and access network types. Because of the dispersion and mobility of fog nodes, wireless communication is critical in this environment. Using wireless networking allows fog communication structures to be flexible, mobile, and reachable. Wi-Fi support at a fog node will rely on a wide range of factors, including the fog node’s role and location in a network hierarchy, as well as its coverage and range. For devices and networks with limited resources, LPWAN (low-power wide area network) has been developed as a protocol. It covers a large area and consumes very little power and data. In agriculture, LPWAN technology is well suited to the task. LPWAN-based protocols include LoRa and SigFox. For long-distance IoT connectivity, cellular networks are the best option. In con- trast, all mobile phone network technologies need RF licensing, IP protection, and energy requirements, all of which add up to a substantial price tag. Communication options for the IoT that use the NB-IoT and LTE M protocols are aimed at providing low-power and low-cost IoT communication platforms. IoT connection is predicted to rise because of a new mobile network, 5G. As a further benefit, it promises lower costs, lower power consumption, and lower latency. A wide range of devices are interoper- able with IEEE 802.11, the most extensively used local area network protocol. Its power requirement, high data transfer rate, and intermediate range make it ideal for latency-aware fog applications. It was for this reason that IEEE 802.11ah and IEEE 802.11ax (HaLow) were created. The typical structural system of MAC layer protocols was found to be completely at odds with the needs for IoT low-power and multi-hub connectivity. This specification has been the most popular MAC layer technology for the IoT since its debut (IoT). The low data rate and medium range of Zigbee and 6LoWPAN make them excellent for building automation. Personal area network technologies, such as NFC, Bluetooth Low Energy, and Radio Frequency Identification (RFID), may be useful in wearing fitness equipment, object tracking, and check-in systems (RFID). The size of the fog platforms will be determined by the amount of the region it will cover; many antenna options are available. The live video broadcasting App04 makes use of directional Wi-Fi anten- nae. The many ways an access network may be used for various applications. There are certain programs that are presented more than once because they use various kinds of access networks and deployment tiers. Real-world test beds are also used for some of the applications being examined. Examples include Power Consumption Management, Vehicle Video Processing, and Vehicle Fog Computing, which explore Dedicated Short Range Communication, LTE, and VFC. The computers that make up a fog-computing platform are not only physically diverse but also in terms of their processing, storage, and network bandwidth cap- abilities. Essentially, they are the foundation of fog infrastructures. When compared to standard cloud systems, fog architectures employ modest computational fog nodes Introduction of fog computing 19
- 43. or servers scattered throughout a broad geographic region in order to service a greater number of users. A user’s latency is highly dependent on where the closest fog nodes are located, as they may be placed anywhere between end-users and a data center. Depending on the application’s needs, developers must choose the suitable fog nodes to enhance the application’s QoS. In order to better understand how the surveyed apps are being deployed, we investigate several fog nodes. Comparison of prospective fog nodes is using the following characteristics: In general, fog nodes may be divided into stationary and mobile nodes. Static nodes are considered at strategic locations. Small- scale data centers and personal PCs are examples of “static nodes,” which are nodes that do not change their locations. Because they are essentially immobile, these devices must be permanently installed. Depending on where they are deployed, static nodes may be further classified into subgroups. Examples include base stations, net- working equipment (such as networking equipment), and micro data centers. Setting up and configuring base stations is more challenging, due to their mobility and smaller size. Drones, vehicles, and other node mobility are examples of single-board devices. However, as indicated in this opening paragraph, nodes may be used to consolidate resources longitudinally. Folio nodes serve as the deployment method for our reference apps, providing a scalable and reliable infrastructure for hosting and delivering applications while ensuring efficient resource allocation and management. Depending on the needs, several of the studied apps might be implemented with one or more possible fog nodes (i.e., computation capacity and proximity). Apps concentrate on near proxi- mity: IoT. When computing resources are needed in close proximity to end custo- mers, single-board PCs are often used. Sensor data from the surrounding environment is collected and processed locally for example. Because of their tiny size, single-board computers may be readily installed and relocated. Traffic congestion management, autonomous driving, and other vehicle-based applications typically make use of the vehicle’s built-in computing power to analyze data collected from the roadside. Apps that use drones to move computing resources, such are known as drone-based appli- cations. Using single-board computers, the drone is able to process locally and communicate with other fog nodes in the vicinity. Applications requiring a lot of processing power are often run on laptops or in tiny data centers. 1.7 Distributed with the fog One of the most important aspects of fog computing is its ability to be distributed. If you want minimal latency, you need several nodes spread out over the network, not just one at the network’s edge. All users in a designated region may access adjacent resources thanks to the nodes spread. There are two ways that fog- computing systems may distribute their output: In both equipment and software partitioning, the dispersed nodes and the instances and components of programs are shown. For the distribution of hardware, there are two typical options. It is called horizontal node distribution, and it involves having several nodes on the same layer in the design. There is also vertical distribution, which is used when nodes differ for 20 Enabling technologies for smart fog computing
- 44. the resources they have (e.g., via a hardware update). To better serve a larger number of users, as a rule of thumb, nodes with even more materials tend to be placed in a greater vertical layer, while those with fewer resources tend to be spread out across a larger area; all of these separate levels are geared toward a single goal: a trip to the endless cloud. To distribute an application across a cluster, replication and multicomponent may be employed. Each element is often a microservice hosted on a different node. As an alternative, a wide range of applications may be duplicated over numerous nodes. A key selling factor for many of the programs that wish to run on top of fog is its dispersion. A regular dispersion of fog nodes is needed when there are a significant number of visitors in a particular application. According to the connected systems that cover the street lights, nodes should indeed be built based on the location of these devices. Replication and horizontal dispersion are both required. A number of requests rely on the continuous function and replication of the same components across several nodes. Video stream processing, for example, makes extensive use of this technique due to its high computational demands. The purpose of this replication is to reduce latency and increase performance by distributing the copy over many nodes. Fog clusters are tiered such that the edge nodes gather data, which is then delivered to the fog nodes, which analyze it and only give back the results to the cloud for cloud-based applications that need vertical distribution. In the case of data streams, this design is very economical, since only the output results need to be sent. This is an excellent method for reducing internet latency and traffic conditions. Apps that do not need to be distributed at all tend to be edge-only or cloud-based apps, which can operate on either side of a network. 1.8 Fog service models Fog-computing systems, like cloud computing platforms, allow users to access virtualized resources at various levels of abstraction. Depending on whether they provide infrastructure, platform, or software, we may divide them into three groups. To distinguish them from their cloud-only equivalents, we’ve given them the names Fog-Infrastructure-as-a-Service (FogIaaS), Fog Platform as a Service (FogPaaS), and Fog-Software-as-a-Service (FogSaaS). ● FogIaaS: It is possible to employ various types of hardware, including CPUs, networks, and discs, using FogIaaS. A wide range of operating systems and tools are available to the end-users, giving them complete control over how they utilize the resources. ● FogPaaS: Customers may utilize it in order to get necessary software products and other capabilities for running multiple and developing software; you may use FogPaaS. A company that develops software testing and deployment pro- cesses may be streamlined and cost-effective thanks to FogPaaS. ● FogSaaS: It is possible to utilize software programs without having to install them on your own computer using FogSaaS. Using a web browser, users may access the services over a distant network. Introduction of fog computing 21
- 45. According to the service models they use, the reference apps are categorized. As an example, there is just one app for this. ● Middlewares are essential: Cloud-based FogPaaS services may be needed as more fog-computing apps are created in order to allow simple application development on fog platforms. We’ve compiled a list of the most popular middlewares here. The Big Data group first came up with the idea of data stream processing technologies. There was some interest in a fog-computing environment, which might reduce data transfers between IoT devices and cloud servers. There are a variety of systems out there, each with its own unique set of capabilities. ● Function-as-a-service: Event-driven, serverless applications may be built with the help of the service. Code or functionality may be developed and managed without the need for servers or server administration. IoT devices may be used in a function-as-a-service architecture, in which a sort of cloud is computing. Fog computing is trying to make use of this. In essence, this is edge computing, where gadgets like IoT and smartphones, web applications, and other endpoints are connected to the cloud technologies exist. Computer net- works use message-oriented middleware (MOM) to communicate with one another using MOM. Distributed and heterogeneous components are supported by a software or hardware infrastructure that seeks to provide message recep- tion and transmission. In order to simplify the development of applications for numerous operating systems and network protocols, it has been designed. In fog-computing settings, MOM is utilized to increase the scalability of fog nodes and job scheduling. An ecosystem where apps can operate indepen- dently of what they do is what web application servers offer. Typically, they have a variety of service levels, each of which solves a specific issue. A cloud- based database server can provide a variety of functions, including web page serving, container models or services for applications, manufacturing require- ments, load balancing over several web hardware, and monitoring and imple- mentation tools. Data centers and end devices may benefit from improved management of and program their computing, networking, and storage resources by enhancing the capabilities of application servers in the cloud. The reference applications make use of a variety of middleware. Middleware is included many times since some applications use more than one kind. Unspecified apps are those that do not specify a middleware type. ● Data processing methods: For the fog architecture, it is important to know what kind of data the nodes process in the system. Data volume and processing timeliness are directly linked to this information, which may be found. Cloud computing is generally seen as a dispersed network that may be expanded horizontally to accommodate additional nodes in the event that the current pool of resources is inadequate. The allocation of a single job among several nodes is seldom discussed, despite this fact. It is common for nodes to be able to host a whole job and, as such, to have processing and storage capacity that is tai- lored to the timeliness of application scenarios. It shows how the apps are 22 Enabling technologies for smart fog computing
- 46. organized based on the amount of data they handle. Only a few applications demand a considerable amount of computer power to handle textual data. Delay-sensitive applications may need a lot of time to process sensor data. It is possible that the amount of processing power required will vary based on the kind and quantity of sensors. Even the most demanding programs, it has been found, can handle static graphics or even video in certain cases. Use of spe- cialist hardware, such as graphics processing units, is typically required for this kind of processing. ● Process automation data: Sensor data and process automation have led to an ever-increasing amount of data being generated. The number, size, and scale of data created and kept is the most crucial factor in the Big Data community’s opinion. Fog is seldom used to store large volumes of data, as we have seen. Fog-computing platforms, on the other hand, are often used to handle data streams that need quick processing or filtering before redelivery to the initial recipients. The quantity of data handled by each software ranges from a few kilobytes up to many terabytes. The great amount of individuals deals with extremely little amounts of data, in the range of kilobytes to megabytes in size, on a regular basis. There is a high demand for real-time operations, which necessitates that data be collected, analyzed, and delivered immediately. Data stream or message-oriented algorithms may need higher storage space when many copies are required to provide high availability and high parallelism. Similar approaches are used when dealing with large volumes of camera- generated data. In many cases, cloud storage is required for apps needing long- term and dispersed retention in fog nodes, such as those that use cameras or personal files (e.g., photographs). A company’s customers and IoT devices influence how much data it can store. For applications that use machine learning or deep learning, a large amount of historical data is essential (such as pattern identification from video camera feeds). The significant number of these programs uses private cloud infrastructures that enable managers to set aside a certain amount of storage space for each application. ● Sensitivity to data speed and delay: Fog computing’s reduced latency between users and resources is a key feature. When parts of the resources are near the end-users, then the widely scattered nodes fulfill a function. Applications seeking ultra-low latencies that may not surpass a few milli- seconds regardless of the velocity of the incoming data will be motivated by fog’s promised low latency. In the Big Data community, the term “data velo- city” is used to describe both the rate at which new data are generated and the time it takes to analyze it. As the number of networked devices and systems grows, so does the amount of data that has to be processed. This is especially true in the field of IoT. According to the application, data output might range from a few kilobytes per second to several megabytes per second. There is a wide range of possible reaction times for different applications, ranging from milliseconds to seconds or even longer. These two metrics are used to classify the reference apps. We can see that most of the programs’ input data genera- tion rates are in the MBps or even kBps range. Data production rates of this Introduction of fog computing 23
- 47. magnitude may at first seem manageable, but most applications need a rapid reaction to the data they create. Consequently, any fog-computing platform may struggle to handle an increase in data creation. We’ve discovered patterns in the apps we’ve tested. IoT-based apps don’t need any latency constraints; however, a large number of other IoT-based applications require low latency for optimal operation. “Low latency” might signify many things depending on the application. Latency requirements for certain apps are quite stringent, but other applications may run with less stringent time limits. Making decisions in the present moment: Ultra-low latency is critical for apps. Decisions must be made quickly in order to prevent collisions, for example, on a fog platform for autonomous cars. User experience of the highest standard gaming and video streaming are two of the most common examples of applications in this area. Rather than endanger lives, long response times might have a negative impact on user experience in these types of apps. In order to fulfill data velocity demands, a humongous fog-cloud computing service must be able to analyze model parameters close to where it was originated and only transport pre- processed data to other fog- or cloud-hosted components. ● Multiple sources of information: Data from a variety of different sources must be integrated into a variety of fog applications. A good example of this is App06, which uses drones to deliver items. The recipient’s present location and other critical useful data, such as control temperature or the most energy- efficient methods, may be found on the internet; this may be necessary for this application. In order to aid the application’s decision-making process, certain metrics must be accessed from outside sources since they are seldom acces- sible locally. It’s possible that there is no established trust relationship between these data owners and the corporations that offer them. Fog applications may have to cope with a variety of security protocols (keys, algorithms, etc.) from diverse data providers since there are a variety of people in charge of inde- pendent bodies. There are a variety of data security protocols and methods used by the various providers, which means that the fog application must be able to access and consume data in accordance with these protocols and pro- cesses. Data suppliers are included, which organize our reference apps. Specific procedures for the various applications that rely on several indepen- dent data suppliers may be required for future fog-computing platforms. ● Sensitivity to the importance of privacy: Data are fueling most of the services we use in today’s cultures. Data about users are required in order to utilize the service, on the other hand. While many individuals are worried about how their personal information may be used, others choose to keep their personal infor- mation private. Apps that monitor physical activity on a user’s daily schedule and provide advices on how to live a more healthy lifestyle are an example of this. In spite of their curiosity, the users do not want their daily physical activity to be made public by their neighbors or coworkers. Concerns over European indivi- duals’ privacy prompted the European Commission to issue the General Data Protection Regulation. Legal, technological, sociological, and other approaches may all be used to describe what we mean by privacy. The laws and actions that 24 Enabling technologies for smart fog computing
- 48. may be done to ensure that only the intended recipient receives personal data are referred to as data privacy in this article. Non-authorized parties should not be able to access an individual’s personal information, which is defined as “privacy” under the law. Due to their proximity to the user, fog apps typically gain access to confidential user data. Fog computing, like IoT and cloud computing, has the same privacy concerns. We also deal with data that are ordinarily owned by a user and are located near a fog node in fog computing. Because the position of fog nodes in the vicinity of a user may be determined, the issue of user privacy is made worse. The three layers of data privacy are as follows: ● The data are open to the public. For instance, a city’s street names are open to the public. ● If a number of requirements are met, some of the data may be accessible. Depending on local regulation, a city’s list of dangerous streets could be pri- vate or public. ● The data can only be accessible by a limited number of people or organiza- tions. Personal health information, for example, is often kept private. ● Security awareness: In any large-scale computer system, data security is a major problem. If you’re using fog computing, regardless of whether your IoT data are collected by sensors or is being sent from the cloud, you must assure its privacy and authenticity. Any data transferred should be received at its inevitable con- clusion in the same form that it was sent. As a rule of thumb, sensitive information should only be accessible to the data source and its intended receiver. An IoT device’s health-related data may be of tremendous value to unauthorized organi- zations, making it susceptible to attack. This is a good illustration of the need to maintain strict secrecy. The personal health information of about 80 million Anthem customers was compromised because of this breach in 2005. Encryption is utilized in modern technologies to secure data. Therefore, not all IoT devices are capable of handling the computing needs of cryptography. This is a problem. Second, a fog-computing platform may need to handle the importance of secrecy in the fog/IoT domain because of IoT devices’ uniqueness. Data leaks may be prevented through context-aware security processes, in which the system switches between various degrees of protection based on its immediate surroundings. Because malware has been detected on nearby computers, a node may switch to a stronger encryption technique. It categorizes the data integrity and confidentiality needs of our reference applications. Obviously, a secure environment is preferred by all applications. We only include those programs on our list that pose a sig- nificant risk to their users if their integrity or confidentiality were compromised. 1.8.1 Characteristics of the workload Fog-computing systems must be widely dispersed in order to be near their con- sumers. Thus, they face significant difficulties operating large-scale fog applica- tions, as well as the infrastructure for fog computing. Fog-computing systems must Introduction of fog computing 25
- 49. be able to adapt to dynamic workloads, detect, and correct problematic workloads in order to provide the highest possible performance and high QoS. The workload generated by our reference applications may be divided into two broad groups, each with a handful of subclasses, depending on the peculiarities of their workloads. ● Workload is dynamic and changes based on a variety of factors. ● Fog node position affects the amount of work that must be done. ● Workload fluctuates with the passage of time. ● Depending on how much labor users put in, the workload changes. Component applications are categorized according to the amount of work they can manage at any one time. As far as we can tell, the majority of fog apps are running at a constant load. During dense fog, sensor data are processed by fog apps. As an example, employ security cameras to take frequent photos and send them to the fog for further processing, as shown in the image. Smart cities, industrial automation, intelligent buildings, and smart grids are just a few examples of where these sorts of applications might be used. Dynamic workloads were observed in seven applications: three based on geo- graphic location, one on time, and three based on persons. There is no difference between web-based and cloud-based web services when it comes to different appli- cations. On the other hand, fog-based applications can only be utilized under certain circumstances. The number of neighboring sensors may also vary depending on the location. The number of self-adapting stations may be reduced. Depending on how many other fog nodes are nearby, the fog node may establish a connection. Finally, it catches mobile in real-time transport demands, modifying the load. Fog-computing systems must be designed and operated in a cost-effective way if their workload characteristics are to be understood. For dynamic workloads, fog infrastructure and applications must be created and deployed in a scalable manner. All of these must be included in fog management platforms, including intelligent application distribution and dynamic capital allocation. 1.9 Merits of fog computing ● Limit the quantity of information transferred to the cloud ● Reduces the amount of bandwidth used by the network. ● Enhances the responsiveness of the system. ● Storing information near the edge increases the safety. ● Supports freedom of movement. ● Minimizes latency on the network and the internet. 1.10 Demerits of fog computing ● The cloud’s anytime, anywhere, and data value are diminished by a specific address. ● IP address spoofing and man-in-the-middle attacks are among the security concerns. 26 Enabling technologies for smart fog computing
- 50. ● Identification and privacy considerations. ● Information infrastructure for wireless connections. 1.11 Application of fog computing The IoT makes extensive use of the developing technology of fog computing. The network edge receives data and services from the network core via fog computing. Similar to the cloud, the fog provides data, computation, storage, and enterprise applications to end-users. ● Patient monitoring systems in real-time primary healthcare units are being developed. ● Monitoring of pipelines for leaks, fires, theft, and the like. ● Smart grid control has the ability to switch between various sources of energy. ● Smart farms with agricultural applications and irrigation management systems are being used in agriculture. ● Fleet management and vehicle health monitoring equipment for trucks and buses are also included in this category. ● Automated invoicing and reporting of shopping baskets. ● Fire alarms, temperature control, and intrusion detection are all examples of smart home technology. 1.12 Conclusion Fog-computing applications come in many shapes and sizes, and as a result, the pervasive computing platforms built to serve them must be able to handle needs that span a broad spectrum. Researchers hope that these findings will help future fog platform developers make well-informed judgments regarding the capabilities. They may or may not include the kinds of activities that would most benefit from those inclusions. Fogging is a term used to describe a hardware implementation where data, processing, and applications are centered at the edge of the network instead of being on the internet. Service latencies and customer satisfaction are both improved consequently. Fog computing eliminates the need for a return trip to the cloud for analysis, allowing users to react to events faster. This leads to increased organizational capabilities, better service levels, as well as greater safety for enterprises that use fog computing. References [1] S. Yi, C. Li, and Q. Li, A Survey of Fog Computing: Concepts, Applications and Issues, In Proceedings of the 2015 Workshop on Mobile Big Data, New York, USA, 2015, pp. 37–42. Introduction of fog computing 27
- 51. [2] F. Bonomi, R. Milito, J. Zhu, and S. Addepalli, Fog Computing and Its Role in the Internet of Things, In Proceedings of the First Edition of the MCC Workshop on Mobile Cloud Computing, New York, USA, 2012, pp. 13–16. [3] S. Yangui, P. Ravindran, O. Bibani et al., A Platform As-a-Service for Hybrid Cloud/Fog Environments, In IEEE International Symposium on Local and Metropolitan Area Networks (LANMAN), 2016, pp. 1–7. [4] M. Aazam and E. N. Huh, Dynamic Resource Provisioning through Fog Micro Datacenter, In IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), 2015, pp. 105–110. [5] O. Salman, I. Elhajj, A. Kayssi, and A. Chehab, Edge Computing Enabling the Internet of Things, In IEEE 2nd World Forum on Internet of Things (WF- IoT), 2015, pp. 603–608. [6] H. T. Dinh, C. Lee, D. Niyato, and P. Wang, A Survey of Mobile Cloud Computing: Architecture, Applications, and Approaches, Wireless Communications and Mobile Computing, 13, 1587–1611, 2013. [7] J. M. Kang, H. Bannazadeh, H. Rahimi, T. Lin, M. Faraji, and A. LeonGarcia, Software-defined infrastructure and the Future Central Office, In IEEE International Conference on Communications Workshops (ICC), 2013, pp. 225–229. [8] H. Hejazi, H. Rajab, T. Cinkler, and L. Lengyel, Survey of Platforms for Massive IoT, In IEEE International Conference on Future IoT Technologies (Future IoT), 2018, pp. 1–8. [9] W. Shi, J. Cao, Q. Zhang, Y. Li, and L. Xu, Edge Computing: Vision and Challenges, IEEE Internet of Things Journal, 3, 637–646, 2016. [10] F. Bonomi, R. Milito, P. Natarajan, and J. Zhu, Fog Computing: A Platform for Internet of Things and Analytics in Big Data and Internet of Things: A Roadmap for Smart Environments, Springer, Cham, 2014, pp. 169–186. [11] S. Yi, Z. Qin, and Q. Li, Security and Privacy Issues of Fog Computing: A Survey, In Proceedings of the International Conference on Wireless Algorithms, Systems, and Applications, 2015, pp. 685–695. [12] P. Hu, S. Dhelim, H. Ning, and T. Qiu, Survey on Fog Computing: Architecture, Key Technologies, Applications and Open Issues, Journal of Network and Computer Applications, 98, 27–42, 2017. [13] A. V. Dastjerdi, H. Gupta, R. N. Calheiros, S. K. Ghosh, and R. Buyya, Fog Computing: Principles, Architectures, and Applications ArXiv160102752 Cs, 2016. [14] A. V. Dastjerdi and R. Buyya, Fog Computing: Helping the Internet of Things Realize Its Potential, Computer, 49, 112–116, 2016. [15] C. C. Byers, Architectural Imperatives for Fog Computing: Use Cases, Requirements, and Architectural Techniques for Fog-Enabled IoT Networks, IEEE Communications Magazine, 55, pp. 14–20, 2017. [16] R. Mahmud, R. Kotagiri, and R. Buyya, Fog Computing: A Taxonomy, Survey and Future Directions in Internet of Everything, Springer, Singapore, 2018, pp. 103–130. 28 Enabling technologies for smart fog computing
- 52. [17] C. Mouradian, D. Naboulsi, S. Yangui, R. H. Glitho, M. J. Morrow, and P. A. Polakos, A Comprehensive Survey on Fog Computing: State-of-the-Art and Research Challenges, IEEE Communications Surveys Tutorials, 20, 416– 464, 2018. [18] R. K. Naha, S. Garg, and A. Chan, Fog Computing Architecture: Survey and Challenges, arXiv:1811.09047 [cs], Nov. 2018. [19] A. Yousefpour, C. Fung, T. Nguyen et al., All One Needs to Know about Fog Computing and Related Edge Computing Paradigms: A Complete Survey, Journal of Systems Architecture, 98, 289–330, 2019. [20] M. Eder, Hypervisor-vs. Container-based Virtualization in Future Internet (FI) and Innovative Internet Technologies and Mobile Communications (IITM), Seminars FI / IITM WS 15/16, Network Architectures and Services, 1, pp. 11–17, 2016. [21] D. Zeng, L. Gu, S. Guo, Z. Cheng, and S. Yu, Joint Optimization of Task Scheduling and Image Placement in Fog Computing Sup-ported Software Defined Embedded System, IEEE Transactions on Computers, 65, pp. 3702–3712, 2016. [22] H. Xiang, W. Zhou, M. Daneshmand, and M. Peng, Network Slicing in Fog Radio Access Networks: Issues and Challenges, IEEE Communications Magazine, 55, pp. 110–116, 2017. [23] J. Santos, T. Walters, B. Volckaeert and F. De Turck, Fog Computing: Enabling the Management and Orchestration of Smart City Applications in 5G Networks, MDPI Journal Entropy, 2017, 6, pp. 4–15. [24] D. Zhao, D. Liao, G. Sun, and S. Xu, Towards Resource-Efficient Service Function Chain Deployment in Cloud-Fog Computing, IEEE Access, 6, pp. 66754–66766, 2018. [25] A. Shawish and M. Salama, Cloud Computing: Paradigms and Technologies in Inter-Cooperative Collective Intelligence: Techniques and Applications, Springer, Berlin, Heidelberg, 2014, pp. 39–67. [26] M. Armbrust, A. Fox, R. Griffith, et al., A View of Cloud Computing Commun ACM, 53, 50-58, 2010. P. Mell and T. Grance, The NIST Definition of Cloud Computing National Institute of Standards and Technology and Information Technology Laboratory, 2009. [27] R. Jain and S. Paul, Network Virtualization and Software Defined Networking for Cloud Computing: A Survey, IEEE Communications Magazine, 51, 24–31, 2013. Introduction of fog computing 29
- 53. This page intentionally left blank
- 54. Chapter 2 Fog computing in the IoT environment Abstract Fog computing has emerged as a promising paradigm in the field of Internet of Things (IoT), addressing the limitations of cloud-centric architectures. This chapter provides a comprehensive overview of fog computing, focusing on its background, scope, problem definition, aims, and analysis in the IoT environment. Background: The exponential growth of IoT devices has led to an overwhelming influx of data, posing challenges for cloud-centric architectures in terms of latency, bandwidth, and network congestion. Fog computing, an extension of cloud computing, aims to bring computational resources closer to the network edge, enabling real-time data processing, low latency, and reduced network traffic. Fog computing leverages the proximity and distributed nature of fog nodes to enhance the performance and efficiency of IoT applications. Scope: This chapter explores the key components and architecture of fog com- puting, including fog nodes, gateways, and cloud–fog collaboration models, high- lighting their roles in the IoT ecosystem. Various applications and use cases of fog computing in different domains such as smart cities, healthcare, transportation, and industrial automation are examined to demonstrate its versatility. The integration of fog computing with emerging technologies like machine learning, artificial intelligence (AI), and blockchain is discussed, highlighting the potential for advanced analytics and enhanced security. Problem definition: The limitations of cloud-centric architectures, such as high latency and excessive network traffic, hinder the seamless execution of real-time IoT applications. Centralized cloud processing raises concerns regarding data privacy, as sensitive information may be transmitted and stored in distant data centers. The resource-constrained nature of edge devices in IoT environments necessitates an effi- cient and scalable computing approach to handle the increasing computational demands. Aim: The primary aim of this chapter is to provide a comprehensive understanding of fog computing’s principles, mechanisms, and benefits in the context of the IoT environment. By highlighting the advantages of fog computing, this study aims to encourage the adoption of fog-based architectures for overcoming the limitations of cloud-centric approaches. The chapter aims to identify the potential challenges and
- 55. Random documents with unrelated content Scribd suggests to you:
- 56. Storjunkar with his: these are drawn on the top of the line; after this they draw another line parallel to the former, only half cross the drum, on this stands the image of Christ with some of his Apostles. Whatever is drawn above these two lines represents birds, Stars, and the Moon; below these they place the Sun, as middlemost of the Planets, in the very middle of the drum, upon which they put a bunch of brazen rings when they beat it. Below the Sun they paint the terrestrial things, and living creatures; as Bears, Wolves, Rain-dears, Otters, Foxes, Serpents: as also Marshes, Lakes, Rivers, &c. This is the description of the drum according to Sam. Rheen, of which this is the picture.
- 57. The Explication of the Figures. In the Drum A. a markes Thor. b Thors Servant. c Storjunkare. d his Servant. e Birds. f Stars. g Christ. h his Apostles. i a Bear. k a Wolf. l a Rain-deer. m an Ox. n the Sun. o a Lake. p a Fox. q a Squeril. r a Serpent. In the Drum B. a denotes God the Father. b Jesus Christ. c the Holy Ghost. d S. John. e Death. f a Goat. g a Squeril. h Heaven. i the Sun. l a Wolf. m the fish Siik. n a Cock. o Friendship with the wild Rain- deer. p Anundus Eerici (whose Drum this was) killing a Wolf. q Gifts. r an Otter. s the friendship of other Lapps. t a Swan. u a sign to try
- 58. the condition of others, and whether a disease be incurable. x a Bear. y a Hog. β a Fish. γ one carrying a Soul to Hell. I have observed that severall of their drums have not the same pictures upon them, I have three very different; one, which is here set down, marked by the letter B. They are described differently by Tornæus, in wch the figures are distinguished so as to refer to several places, of which there are chiefly three. In the first stands Norland, and other Countries of Sweden, which are placed on the South side of the drum, and are separated by a line from the rest; in this also is contained the next great City, where they trafic most; as in the drums made at Torne, or Kiemi, there is drawn the City Torne, with the Temple, Priest, and Governour of the Laplanders, and many others with whom they have any concerns: as also the highway that lies betwixt them and Torne, by which they discover when their Priest, or Governour will come; besides other affairs managed in those parts. On the North part, Norway is described with all that is contained in it. In the middle of these two stands Lapland, this takes up the greatest part of the drum: in it are the several sorts of beasts that are in the Countrey, here they picture herds of Rain-dears, Bears, Foxes, Wolves, and all manner of wild beasts, to signifie when, and in what place they may find them. If a tame Rain-dear be lost, how they may get him againe. Whether the Rain-deers young ones will live. Whether their net fishing will be successfull. If sick men will recover, or not. Whether women great with child shall have a safe delivery. Or such, or such a man will die of such a distemper, or by what other; and other things of the like nature which they are desirous to know. I cannot give an account of the reason for this difference in the drums, unless it is that some of them are made for more malicious designs, others again for each man’s private purpose. Upon this account I believe, according to the nature of the business they intend, they add, and blot out, and sometimes wholly change the figures. But that you may the better understand the diversity of the drums, here are two represented to you, both which I had out of the Study of the Chancellour of the Kingdom.
- 59. The explication of the Figures. In the Drum C. a denotes Birds. b black Foxes. c Tinur, a God. d Thor, a God. e Thors hammer. f Storjunkare. g a wooden Idol. h his Servant. i a Star. k an Ox. l a Goat. m a Star. n the Moon. o the Sun. p a Star. q another Star. r a Wolf. The two greater Figures represent, one the upper, the other the lower side of the Drum, and so do also the two lesser.
- 60. Besides these two drums, I had also a third given me by the same Lord of as great a size as any that can be usually met with.
- 61. To these I add a fourth, given me by the Illustrious Baron Lieutenant Henry Flemming, mark’t with the letter F.
- 62. Now there are two things required to fit the drum for use, an Index and a Hammer, that shews among the pictures the thing they enquire after, with this they beat the drum. The Index is the bunch of brazen rings mentioned before. They first place one great ring upon the drum, then they hang severall small ones upon that; the shape of the Index’s is very different, for of these I have one made of copper, of the bigness of a Dollar, with a square hole in the middle, several small chains hanging about it instead of rings. Another hath an Alchymy ring, on which a small round plate of copper is hung by little chains. I have seen another also of bone, in the shape of the Greek Δ, with rings about it; and others of a quite different make. I have described mine under the drums A, and B, by the mark G; but the common sort of rings are of copper, and those upon the Chancellors drums are altogether such. Some writers call these rings serpents, or brazen frogs, and toads, not that they resemble them, but because by them they signifie these creature, whose pictures they often use in their conjuring, as supposing them very grateful and acceptable to the Devil. The Laplanders call the Index Arpa, or Quobdas; and make it indifferently of any sort of metal. The hammer they use in raising their familiars, is not the Smith’s; which was the errour of him that drew it in Olaus Magn. but is an instrument belonging only to the Laplanders, and called by a peculiar name by them: it is made of a Rain-deers horn, branching like a fork, this is the head of the hammer, the other part serves for the handle. The instrument is placed under the two drums A. B. with the letter H, with the hammer they beat the drum, not so much to make a noise, as by the drumming to move the ring lying on the skin, so as to pass over the pictures, and shew what they sought after. This is the description of the drum, with all its necessaries as it is used by the Laplanders that are subject to the Swedes; the Finlappers also that are under the Crown of Danemarke, make use of drums something different in fashion from the former; yet however the difference is so small, that I believe their drums are not of a different kind from ours, but made only for some particular uses. I shall give an account of one of those, described in Wormius’s Study, who saies that “the Laplanders drum, which they use in their magic, and by beating which they discover those things they desired, is made of an oval piece of wood hollowed, in length a foot, in breadth ten
- 63. inches; in this they make six holes, and put a handle to it, that they may hold in the left hand, whilst they beat it with the other; upon it they stretch over a skin, painted with diverse rude figures, drawn with blood, or red; upon this lies a piece of brass, in the shape of a Rhomboides, somewhat convexe, about two inches in diameter, in the middle of this, and at each corner hangs a small chain. The instrument, with which they beat the drum, is of bone, six inches long, about the thickness of a little finger, and made much like the Latine T.” This instrument the Laplanders use for diverse designs, and are of opinion that whatever they do it is don by the help of this. For this reason they have it in great esteem and reverence, taking such care in securing it, that they wrap it with the Index, and hammer, up in a Lambskin, and for its greater safety, lay it in some private place. But I think it an errour, to suppose them to lay it in a Lambskin: for it is written in some places Loomskin, which signifies the skin of a bird that lives altogether in the water. They think it so sacred, and holy, that they suffer no maid that is marriageable to touch it; and if they remove it from place to place, they carry it the last of all, and this must be don too only by men; or else they go with it thro some untrod way, that no body may either meet or follow them. The reason they give for their great care in this particular, is, because they believe if any one, especially a maid that is marriageable, should follow the same way, they would in three daies time at least fall into some desperate disease, and commonly without any hopes of recovery. This they seem to verifie by many examples, that we may give the more credit to it; and we have the less reason to doubt the truth of this, since the devil severely commands his worship to be observed, and suffers not those rites and customs he hath imposed to be violated, so long as God is pleased to grant him this liberty. Now because it may happen sometimes that a woman may out of necessity be constrained to go that way, by which the drum hath bin carried, the devil is so favorable as to permit it without any danger, upon condition she first offers a brazen ring to the drum.
- 64. In the next place, because they believe they can effect very strange things by the drum, we will shew what they are, and the
- 65. manner used to perform them. These are three, belonging either to their hunting, their sacred affairs, or lastly the enquiring into things far distant. I find four chiefly mentioned by another Writer, the first is, the knowing the state of affairs in forreign Countries. The second, what success their designs in hand will meet. With the third, how to cure diseases. The fourth, what Sacrifices their Gods will be pleased to accept, and what beast each God desires or dislikes most. As to the way in making enquiries, it is not the same among all these artists. But the great thing they generally observe, is, to stretch the skin very stiff, which is don by holding it to the fire. The next is, that they beat not altogether in the same place, but round about the Index; then that they beat softly at first, presently quicker, and continue this till they have effected their intent. The drummer first lifts up the drum by degrees, then beats softly about the Index, till it begins to stirr, and when it is removed some distance from its first place to either side, he strikes harder, till the Index points at something, from whence he may collect what he sought for. They take care also that as well he that beats the drum, as those that are present at the ceremony, should be upon their knees. As to the occasions of their beating thus, the later of those is already discoursed of. Now we proceed to the rest, the first of which is concerning their enquiries into things acted in remote parts. Those who desire to know the condition of their friends, or affairs abroad, whether distant five hundred, or a thousand miles, go to some Laplander, or Finlander skilfull in this art, and present him with a linen garment, or piece of silver, as his reward, for satisfying them in their demands. An example of this nature is to be seen upon record, at Bergen, a famous Market Town in Norway, where the effects of the German Merchants are registred; in this place there was one John Delling, Factor then to a German, to whom a certain Finlapper of Norway came with James Samaousuend: of him John Delling enquired about his Master then in Germany; the Finlapper readily consenting to tell him, like a drunken man presently made a great bawling, then reeling and dancing about several times in a circle, fell at last upon the ground, lying there sometime as if he were dead, then starting up on a suddain, related to him all things concerning his Master, which were afterwards found to agree to what he reported. There are many more instances of this kind: the most considerable, is one concerning a
- 66. Laplander, now living, who gave Tornæus an account of the Journey he first made to Lapland, tho he had never seen him before that time; which, altho it was true, Tornæus dissembled to him, least he might glory too much in his devilish practises, and rely upon them, as the only means whereby he might attain to truth. The autority of this man is so considerable, that it may gain credit enough to the Story. As to the method taken in making discoveries, it is very different. Olaus Magn. describes it thus, the drummer goes into some private room, accompanied by one single person, besides his wife, and by beating the drum moves the Index about, muttering at the same time several charms, then presently he falls into an extasie, and lies for a short time as if dead; in the mean while his companion takes great care, that no gnat, flie, or other living creature touch him; for his Soul is carried by some ill Genius into a forreign Countrey, from whence it is brought back with a knife, ring, or some other token, of his knowledg, of what is done in those parts; after this rising up, he relates all the circumstances belonging to the business that was enquired after; and that they may seem certainly so, he shews what he hath brought from thence. Petr. Claud. makes no mention either of the drum, charms, company, or those things he brings with him; but saies he casts himself upon the ground, grows black in the face, lying as if dead for an hour or two; according as the distance of the place is, of which he makes enquiry; when he awakes he gives a full account of all affairs there. It is clear from what was said before, that they made use of a drum; and ’tis observed that for this sort of conjuring the lower part of the drum, whereby they hold it, was commonly shaped like a cross. One of this make was given me by the Lord Henry Flemming, Colonel of a foot Regiment in Finland, the Figure of it is in the page foregoing. They hang about it several claws, and bones of the creatures they take. That several persons also, as well men as women, are permitted to be present at this ceremony, is asserted by Sam. Rheen in his history, where he saies that the drummer sings a song, called by them Joiike, and the men and women that are present sing likewise, some in higher some in lower notes, this they call Duura. Next as to the casting themselves on the ground, there are various relations, some think them not really, but only in appearance dead; others are apt to believe that the soul departs from the body, and after its travell abroad,
- 67. returns again. But without doubt this is false, for is it impossible, for either man, or devil, to restore the soul to the body it hath once left. So that I believe the devil only stifles the faculties of the soul for a time, and hinders their operations. Now after the drummer falls down, he laies his drum as near as possibly on his head, in this posture.
- 68. Those in the mean time that are present, leave not off singing all the time he lies sweating in this agony; which they do not only to put
- 69. him in mind, when he awakes, of the business he was to know; but also that he might recover out of this trance, which he would never do, (as they imagine) if they either ceased singing, or any one stirred him with their hand or foot. This perhaps is the reason why they suffer no flie, or any living creature to touch him; and it is upon this account only that they watch him so diligently, and not out of any fear they have least the devil should take away his body; which opinion of Peucers is altogether false. It is uncertain how long they lye in this manner, but it is commonly according as the place where they make their discovery, is nearer or farther off; but the time never exceeds 24 houres, let the place be at never so great a distance. After he awakes he shews them some tokens to confirm their belief in what he tells them. This is the first and chiefest use they make of the drum. The next is, how to know the event of their own concerns, and what success their hunting will have, or any other business which they undertake, for they seldom venture on any thing, without first consulting that. In order to the knowing this, they place the bunch of rings on the picture of the Sun in the drum; then they beat, singing at the same time; if the rings go round towards the right hand, according to the Suns course they promise to themselves good health, fortune, and great encrease both of men and beasts; if contrary, towards the left, they expect sickness and all the evils attending on ill success. We may easily ground this opinion of theirs upon the other mentioned above, where they believe the Sun the only Author of all productions. Wherefore when the Index moves according to his motion, it portends prosperity by following his course, from whom they expect all the good they receive. This is the way they take in all their more weighty affairs, as in a journey, hunting, removing their habitations, or any such like thing, of which something before, and more hereafter. Before they hunt they make particular observation which way the Index turns, whether East, West, North, or South; and collect from thence where their game lies. Other things for which the drum is serviceable, are, first, the discovering the nature of diseases, whether they arise from any disorder in the body, or are caused by magic; this being known, then to find the remedy for them, which is commonly by sacrifice to one or other of their angry Gods, but chiefly to Storjunkar, who bears
- 70. greatest autority among them, and if not appeased, leaves them small hopes of recovery. Wherefore the sick person vows a sacrifice, either of a Rain-deer, Bull, Goat, or Ram, or something of this kind to one of the Storiunkars, that stands upon the mountains. The sacrifice is not left to the disposal of the sick man, but must be made according to the directions of the drummer; for he is supposed to be the only man able to advise them in this case, he first discovers which of the Gods is displeased, and what sort of sacrifice is most acceptable to him, for they refuse several, and the same also at several times. But before the drummer appeases their Gods, they give him a copper and a silver ring, putting them on his right arm, then he begins a song, and beats the drum, and all that are present joyn with him in a Chorus; after this according to the place, to which the Index points, he directs them. These are the things commonly done by the drum. The last thing for which they think it necessary, is, the accomplishing their wicked designs, as impairing mens health, or depriving them of their lives; which is frequently enough practised among them, tho not altogether so publicly as heretofore. Some of them account this only unlawful, and exclude themselves out of the number of those, which use it, thinking the other uses of the drum to consist chiefly in doing good. But however this mischievous Art continues still too much among them. Several inhabitants of Kiema in Lapland were apprehended in the year 1671, with drums, for this purpose so large, that they could not be removed from thence, but were burnt in the place. Among those Laplanders there was one four score years of age, that confessed he was bred up in this art from his childhood, who in 1670 upon some quarrell about a pair of mittens, caused a Boar of Kiema to be drowned in a Cataract, for which he was condemned to die, and in order to that was to be carried in chains to the next town in Bothnia, but in the journy he contrived so by his art, that on a suddain, tho he seemed well, and lusty, he died on the sledge, which he had often foretold he would sooner do, then fall into the Executioners hands. As to the ceremonies used in this particular, either in their words, gesture, or any other thing, I can give no account, finding none in those writings, from whence I collected the rest. The reason for this, I suppose, is, because they themselves keep this secret, as the great
- 71. mystery in their art; or that no one would enquire into them, least they should be thought guilty of this damnable sin. Having treated largely of the drum, we come to the other parts of this art, to which also belong proper sorts of instruments: the first is a cord tied with knots for the raising of wind. They, as Zeiglers relates it, tye three magical knots in this cord; when they untie the first, there blows a favorable gale of wind; when the second, a brisket; when the third, the Sea and wind grow mighty stormy, and tempestuous. This, that we have reported concerning the Laplanders, is by Olaus Magnus, and justly, related of the Finlanders, who border on the Sea, and sell winds to those Merchants that trafic with them, when they are at any time detained by a contrary one. The manner is thus, they deliver a small rope with three knots upon it, with this caution, that when they loose the first, they shall have a good wind, if the second, a stronger, if the third, such a storm will arise, that they can neither see how to direct the ship, and avoid rocks, or so much as stand upon the decks, or handle the tackling. No other Writers mention this concerning the Laplanders, and I am apt not to think it at all probable, since they live in an inland Country, bordering no where upon the Sea. Wherefore this properly belongs to the Finlappers in Norway. Now those that are skilled in this art, have command chiefly over the winds that blew at their birth; so that this wind obeys principally one man, that another, as if they obtained this power when they first received their breath; now as this belongs chiefly to the Finlappers and Finlanders of Norway, so doth the stopping of the course of ships, which is altogether of the same nature. This is also attributed to the Laplanders, who according to the different affection they have for Merchants, make the Sea either calmer, or more tempestuous. We come now to their magical Darts, which they make of lead, in length about a finger; by these they execute their revenge upon their enemies, and according to the greatness of the injury received, they wound them with cankrous swellings, either in the arms, or legs, which by the extremity of its pain, kills them in three daies time. They shoot these darts to what distance they please, and that so right too, that they seldom miss their aim. Olaus Magnus reports the same in his writings, which I believe is only a transcript of Zeigler’s, the words
- 72. being the same, and without doubt he follows him in this particular as he hath in many others. But I suppose they are both mistaken, and misrender’d them leaden darts, since I can find no person in these times that knows of any such; neither is there any mention made of them in any other writers, or by the common People, who seldom omit such circumstances as these in their relations. But they might perhaps be mistaken in supposing them to be made of lead, by misunderstanding the word Skott, which is commonly used for their explanation. For when either man or beast is suddainly taken with a disease, by which their strength fails, and they immediately perish; the common People call this that takes them so Skott, that is a dart. This might make Zeigler think to be really some dart, which the inhabitants are wholly ignorant of, and most among us believe these things to be effected by some other means. Petrus Claudius calls it a Gan, which they send abroad: he likens it to a flie, but saies it is some little devil, of which the Finlanders in Norway that excell most in this art, keep great numbers in a leathern bag, and dispatch daily some of them abroad. Of these he relates a story, that happened in his time: an Inhabitant of Helieland, who is still alive, going towards the mountains in Norway to hunt Bears, came to a cave under the side of a hill, where he found an image rudely shapen, which was the Idoll of some Finlander; near this stood a Ganeska, or magical satchel: he opened this, and found in it several blewish flies crawling about, which they call Gans, or spirits, and are daily sent out by the Finlanders to execute their devilish designs. But he seems to intimate no more by this word Gan, then that very thing which endangers mens health, and lives. For he saies that these Finlanders cannot live peaceably, except they let out of their Ganeska or Gankiid, which is the satchel, every day one of the Gans, that is a fly or devil. But if the Gan can find no man to destroy, after they have sent him out, which they seldom do upon no account at all, then he roves about at a venture, and destroies the first thing he meets with; sometimes they command it out to the mountains, to cleave rocks asunder: however these conjurers will, for very trivial causes, send out their Gan to ruine men. This word Gan signifies no more then what Zeigler meant by his dart, for the term by which they express its going out is de Skiuda deris
- 73. Gan, that is, he as it were shoots out his Gan like an arrow, for Skiuda is only proper to the shooting out of an arrow. This is the third thing belonging to their magic, which they use as well against one another as strangers; nay sometimes against those that they know are their equals in the art. Of this kind there happened a notable passage betwixt two Finlanders, one of which was called Asbioern Gankonge, from his great knowledge in the art, the other upon some small difference concerning their skill, or some such trifle, would have destroyed Asbioern, but was still prevented by his too powerfull art, till at last finding an opportunity, as Asbioern lay sleeping under a rock, he immediately dispatcht away a Gan, that cleft the rock asunder, and tumbled it upon him. This happened in the time of Petrus Claud. not long before he wrote his History. Some of the Conjurers are contented only with the power to expell that Gan out of men, or beasts, which others send. This is remarkable among them, that they can hurt no man with their Gan, except they first know his parents name. Now all that the Finlanders and Finlappers of Norway effect by their Gan, the Laplanders do by a thing they call Tyre. This Tyre is a round ball, about the bigness of a wallnut, or small apple, made of the finest hair of a beast, or else of moss, very smooth, and so light that it seems hollow, its colour is a mixture of yellow, green, and ash, but so that the yellow may appear most. I had one of these given me by Mr John Otto Silverstroem, Warden of the Colledge belonging to the metals, and Master of the Mines at Saltzburg and Frahlune. This is the figure of it.
- 74. This Tyre they say is quickened and moved by a particular art; it is sold by the Laplanders, so that he that buies it may hurt whom he pleases with it. They do perswade themselves, and others, that by the Tyre they can send, either Serpents, Toads, Mice, or what they please into any man, to make his torment the greater. It goes like a whirlewind, and as swift as an arrow, and destroies the first man, or beast, that it lights on, so that it often mistakes. Of these we have too many instances in this time, which are too long to insert here: having therefore done with all, or at least the chiefest matters concerning their sacred, and superstitious rites, or worship; we proceed to other affairs.
- 75. CHAP. XII. Of the Government of the Laplanders. We come now to their secular affairs, which are either public or private: we will treat first of the public, to which belong the form and constitution of their Government. This in former times, before they were named Laplanders, was in this manner; they were subject to no neighbouring Country, but were governed among themselves yet so as to be subject to a King, they chose out of their own Nation. Most of them, or at least those which bordered on Norway, and dwelt near the Sea, were under this kind of Government, in the time of Harauld Harfager King of Norway, cotemporary with Ericus the Conqueror, King of the Swedes, this was 900 years after Christ; he conquered the greatest part of Norway, except these Finlanders. The King that reigned over them at that time, was named Mottle. This account was questionless taken from Haralds expedition into Biarmia, and his ruining all that Countrey, except the part belonging to these Finlanders. In those times the name of Laplanders was neither used, nor known, as I have shewn elsewhere, but they retained that of their ancestours, which was also common to all of the same extraction. Their condition was not much altered, after that they took this name; which was when they first sent out Colonies into the inland Countries, on the farther part of the mountains, which divide Swedland from Norway. For they that went out had certainly some Leader, whom without doubt they chose for King, after they had taken possession of those Countries; and I believe they would scarcely submit to any other power whilst that he was living; and this seems the more probable, because no one in those daies would undertake the conquest of a company of poor beggarly fugitives,
- 76. who dwelt among Woods and Deserts, in continual snow and the greatest extremity of cold. This was the Moscovites opinion of them, who tho they dwelt near them, scarcely knew their nature and disposition, and thought it madness to set upon them with a small party, and an adventure of little profit, and less honour to raise an Army against a Country already distressed by poverty. For this reason the Laplanders enjoied their own customs for a long time. The first King of Sweden that had any thoughts of conquering them was Ladulaus the great, who florished about the year 1277, who because it seemed difficult to bring them under the Crown of Sweden, promised those that would undertake the conquest, the government over them. He thought it too expensive to make a public war upon them, when they were to be dealt with as wild beasts; yet however could not endure that a neighbouring People, dwelling almost in the heart of his Country, for they possessed at that time as far as the Bay of Bothnia, should refuse obedience to his Kingdom. Wherefore he thought upon the before mentioned project, and proposed great advantages to private persons, upon which the Birkarli, their neighbours, readily engaged themselves, and effected their enterprize no less successfully. In this design, the plot of a particular person was most remarkable, as is related by Ericus, and recorded by John Buræus. One single man of the Birkarli went towards Lapland to way-lay the Laplanders in their return from Birkala, (at this time no one inhabited on the North side of that allotment) and ordered his wife to cover him over with snow, in the middle of the way where the Laplanders must necessarily pass over him. They came in the night time, and by their passing over him he knew there were fifteen, which were the chief among them, and to whom the rest were in subjection; when they were gone, he immediately arose out of the snow, and going some shorter way, set upon them at unawares, as they passed by, one by one, which is their usual way in travelling, and slew them one after another. None of those that followed perceived the first men slain, it being in the night time, and each of them at some distance from the others; till the last man finding his fellows killed, made a stout resistance, but the Birkarla by the assistance of his wife got the victory, and slew
- 77. him likewise. Thus the most powerfull of them being slain, the rest readily submitted. Some think the Birkarli deluded them by a pretended truce, and that before it was expired, they assaulted them, not suspecting then the least danger, and killing several, subdued the Countrey, as far as the Northern and Western Oceans. We may easily collect from the truce mentioned here, that before their subjection to the Swedes by the Birkarli, there was some kind of war betwixt both: besides, it was shewn above, that Ladulaus could not bring them under his Crown. This perhaps may be Zeiglers meaning, when he describes them as a warlike People, and free for a great time, that they also withstood the Arms of Norway and Sweden, till they were forced at last to yeild; but what Zeigler imputes to their valour, proceeded only from the contemt they were then in, as is plain from the opinion the Moscovites gave of them. And there is little reason to suppose the Swedes were not of the same, since they were overcome only by the allotment of Birkala; and Ladulaus did not conquer them out of any fear he conceived of their forces, but by sleight, foreseeing the small advantages he should receive would not quit the charges of an Army. Thus the Laplanders were brought in subjection by the subtilty and expence of private persons. About the year of our Saviour 1277, the Birkarli had the autority over them; yet so as to acknowledg their dependance on the King of Sweden. Now whether all of them were thus overcome, as those that lived beyond the mountains of Norway, near the Sea, which are the Finlanders, or Lappofinni, is still in doubt, except we collect it from this, that all from the Northern and Western Oceans were certainly subjected. But whatever dispute may arise concerning that, it is manifest the Swedes were the first Conquerours of Lapland, but afterwards the Norwegians and Moscovites following their example, put in also for a part; thus they became subject to these three severall Princes. But to pass by the others, the Swedes enjoyed, for some former ages, half the dominions from Tidisfiorden to Walangar, over the Lappofinni, or maritime Finlanders. This was given by Charles the IX, in his instructions to his Embassadors, sent to the King of Danemark, wherein he made it appear that the Swedes had from former times,
- 78. till then, enjoied half the rights, both sacred and civill, whether as to tributes, punishments, men, or fisherie, with the Crowns of Danemark and Norway. But the Swedes kept only a third part from Malanger to Waranger, those of Norway and Moscovy laying claim to the other two, till in the year 1595, the Moscovites, by a League, delivered up their part, but the Swedes alwaies possessed the mountainous and more neighbouring places from Ladulaus’s time, for near four hundred years, and exercised their autority over them. The Government after the conquest was in the hands of the Birkarli, according to the grant given them by Ladulaus, who ruled over those that dwelt near the Bay of Bothnia, imposed taxes, trafficked with them, and received all the profit of the Salmon fishing, and all other advantages arising from them; but in acknowledgement to the King, as Supreme, they paid a certain number of gray Squirrils skins. The Laplanders, by common consent, received and honored the Bergchara, that is men of the mountains, or Birkarli, as their Governours, and paid them very rich skins, and severall sorts of fish, both for their tribute to the King of Sweden, and their own proper uses. Neither were there any other commissioned by the King in those times to govern them, as will appear afterwards. He, that was their Governor was honored by them with the title of King, his autority was confirmed by the Crown of Sweden, he wore a red robe, as the token of his Roialty; now from this sort of garment, by which the Birkarli were distinguished from others, it is evident they were the first rulers in those parts; and perhaps only one governed them, whilst they dwelt near the Bay of Bothnia, but when they enlarged their possessions farther into the Land, and were divided into severall Counties, each division had its particular Governor. And that it was so, is manifested from the Letters of Gustavus the first, where he divides the Birkarli into Luhlians, Pythians, and Tornians, over which accordingly there were severall Governors. It may perhaps now be a dispute, who these Birkarli were, by whom the Swedes subdued Lapland; Buræus saies they were the Inhabitants of the allotment, of Birkala, but Olaus Magnus is of a different opinion, and calls them Bergchara, that is, men of the mountains, from Berga mountain, and Charar or Karar men. What grounds he
- 79. hath for this, he neither declares, nor can I easily imagine. But I think them so small that they will find little credit any where; for from whence, or from what mountains should they be thus called? not from those of Norway, when at that time no body inhabited there; neither are there any other mountains besides these, from whence they should take this name: moreover, the Birkarli were subjects to the Swedes, and conversed commonly with the Laplanders. The public records also contradict this opinion, for in them there is no mention of Bergcharli, but Birkarleboa. It is yet clearer also from the Letters of Cnute Joanson, written in Latine, in the year 1318, where he saies in the Parliament held at Telge, betwixt the Helsingers and Birkarleboa in his presence, there was issued out this Placart, &c. This serves to confute Olaus. It is more evident that they came from Birkala, an allotment in Tavastia, and described in the Mapps. Next, as to Gustavus the first mentioning the Birkarli, in the foresaid Letters, as belonging to severall marches, viz. Luhla, Pitha, and Torna it was upon this account: the Birkarli that descended from those of Tavastia, were placed in these severall Towns to govern the Laplanders, and because they only had the priviledge of commerce with them, they were called Merchants. They were used in the Summer to buy those commodities of the Merchants that came to Bothnia, which were necessary for the Laplanders, and in the Winter, when the Rivers and Lakes were frozen over, they carried them up into the Countrey. This way of trafic was used by all the Inhabitants of Bothnia, but perhaps only at first by one allotment, which growing populous, severall of the Inhabitants removed farther into the Countrey, and retained the same priviledge that was first granted by Ladulaus, viz. that no one, but they, should claim any priviledges over the Laplanders, either as to the Government, tribute, commerce, or any thing of this nature, which priviledges they for a long time enjoied, as is confirmed by the Letters wrote by Cnute Joanson, in the time of King Smecke, in which it was provided that the Birkarli should not be molested either in their passage to or from the Laplanders. This priviledge they maintained till Gustavus the first, who made a Contract with them at
- 80. Upsal on the 4th of April 1528, concerning the yearly tribute they were to pay to the Crown, for the great advantages they received from the Laplanders. This tribute was only in respect of the priviledges the Birkarli had from Ladulaus’s time till then, these were so largely granted, that they setled them as hereditary upon their children, and none but those descended from the Birkarli could enjoy them. This Gustavus also confirmed according to the former grants made to their ancestors, but with this alteration that they should pay half as much more, as they did formerly. This Government the Birkarli exercised over the Laplanders which they got by subtility, had their autority from the King of Sweden, preserved it in their own family, and delivered it down to their children for near 300 years, till Gustavus the first, by reason of their insulting over the common People, deprived them of this state; for when their riches encreased they oppressed the poorer sort, and extorted so much from them that they left them very little, but that which was worth nothing. Upon this, complaint was made to Gustavus, who thereupon committed Henricus Laurentii to prison, and confiscated most of his estate, taking then the tribute from the Laplanders into his hands, and granted to all People free trading with them. This Henricus Laurentii was without doubt in that time the head of the Birkarli, and I believe the brother of David Laurentii, who, together with Jonas Nicolas, concluded the Treaty with Gustavus in the name of the Birkarli, in the year 1528, for setling the tribute, and other affairs. From hence we may collect they lost their priviledges, not long after this Contract; now it was not only just to deprive them of those priviledges, which they abused in oppressing others, but prudent, as well from the jealousy of too great a power granted to private persons over so large and populous a part of the Kingdome, as out of consideration of its wealth, which was more necessary to the Kings, for driving out the common enemy, ane establishing the Kingdomes liberty, then to maintain the pride of the Birkarli, who besides their injustice, were inconsiderable both in number and strength. Gustavus the first having thus deposed the Birkarli, sent Deputies to gather the tribute, and manage all things in the Kings
- 81. name; the Deputies are called by the Swedes, Lappfougder, by the Laplanders, Konunga Olmai, that is the Kings men; of these there is mention made in the patent granted by Gustavus the first to Mr Michael, the first Priest in Lapland in 1559, the words are to this purpose, We commend all the Inhabitants of Lapland, as well Deputies, as others, &c. These had at first the charge of all public affairs, as will appear in the following Chapter, as for collecting taxes, as executing justice among them. But afterwards, when Charles the ninth divided the Countrey into several parts, and formed it into better order, more were added to the former, for examining causes, convicting of criminals, and other such like things, till at last the state of Government was little different from what it is now. Next under the King, they have a Provincial Judge called by the Swedes, Lagman, under him one of the Senators, Underlagman, next an Interpreter of the Laws, Laglæsaren, and divers others which enquire into causes, and do justice; then they have a Governour of the Province, Landzhœfdingh, a head over the Laplanders, Lappafougten, their Officers who perform all other duties. In this manner the Laplanders are now governed by the Swedes.
- 82. CHAP. XIII. Of the Judicatures and Tributes of the Laplanders. After the manner of their Government, and the discipline they live under, we descend to those affairs that are managed by it; which belong either to the Courts of Judicature, or to the Tribute. I can scarce find any mention of the former. Their own Kings, when they were a free Nation, exercised this autority, and kept the jurisdiction in their own hands; but when the Birkarli ruled them, it depended altogether on their plesure. Zeigler makes no mention of any Judges among them, but saies that if any dispute happened that was dubious, it was referred to the Courts in Swedland; I suppose he means the more weighty controversies, which the Birkarli could not, or did not dare to decide. But these were very rare with them, for great crimes, as theft, rapine, murder, adultery, or such like are seldom committed and scarce known by the Laplanders. They neither borrow nor lend mony, being content with what they possess of their own, which are commonly the occasions of quarrels in other Nations, and maintain so many Lawyers. The chief sin they are guilty of is their magical superstition, which since their embracing Christianity, is forbidden by the Laws, and is not so frequent as formerly. After that Gustavus the first had deposed the Birkarli, and given them Governors of their own, they lived under better discipline, and greater diligence was used in seeing Justice done, but Charles the ninth was the first that took care to have them instructed in the Swedish Laws, and that they should regulate themselves accordingly. This charge was given by the same King in his instructions to Laurentius Laurentii, Governor of Lapland, dated
- 83. from Stockholm on the 10th of Oct. 1610, wherein he commanded him to govern those of Uma, Pitha, and Luhla, according to the Swedish Laws, and to protect them from all injuries. There are at present in Lapland three Governors, and as many Courts of Judicature: the first is called Anundsiœense, or Angermansian, the other Uhmensian, Pithensian and Luhlensian, the other is the Tornensian, and Kiemensian. Over these are particular Governors, who in the Kings name pass Sentence, but in the presence of a Judge and a Priest; where it is observable that they added Priests to the Governors, to restrain them from doing injustice by the autority of their presence. Now as to the time when these Courts were called, it is a doubt, but I believe it was at the Fair times, when they met about all public business; this was commonly twice in a year, viz. in Winter and Summer, according to an order of Charles the ninth’s. It is now in January and February. They were held in the same places where they kept their Markets and Fairs, which were determined in each particular County, as will appear by and by. Now we come to the Tribute they paid, which at first was only skins of beasts, paid not by the Laplanders, but the Birkarli, yet only as an acknowledgement of their subjection to the Crown of Sweden. Buræus calls it naogra timber graoskin, graoskin signifies gray Squirrils skins, of which color the Squirrils were constantly in the Winter; timber denotes the number of the skins, which were fourty, tied together in a bundle. It is uncertain how many of these bundles the Birkarli gave, but in the Contract with Gustavus the first, those of Luhla and Pitha were engaged to pay 8, which makes in all 360 skins, besides two Martins skins. Those also of Torne were taxed with the same number; and shortly after this number was doubled, by an agreement made in 1528. But after the Birkarli had lost their priviledges, for the forementioned reasons, and the King received the tax by Commissioners for himself, it is very probable some more alteration were made. In the year 1602 they paid instead of skins every tenth Rain-deer, and one tenth of all their dried fish; which is clear from the commands given by Charles to his Deputies Olaus Burman and Henry Benegtson, at Stockholm on the 22d of July in
- 84. the same year, to require the tribute in this manner, that so the Laplanders might know what and how much they were to pay: for it seems that from Gustavus the first’s time, till then, the Governors used no constant method in raising it, but sometimes demanded skins, at other times other sorts of goods that seemed most necessary for present use; so that by this uncertainty the tribute grew very heavy upon the Inhabitants, and their Governors took occasion from it to exact what they pleased under pretence of the public account, for their own proper uses. Yet this custom continued not long, being thought perhaps too burthensome to the Laplanders, and very prejudiciable to their herds; wherefore it was ordered in 1606, that every one which was then 17 years of age, should pay either two Bucks, or three Does out of their herds of Rain-deers, and eight pound of dried fish; as also every tenth Fawn out of their stock, and every tenth tun from their fishery. This tax was also imposed on the Birkarli that had any trafic with them. This order was kept a long while, and renewed again by the same King in 1610. The tribute they pay at this time is either mony, Rain-deers, or skins, either plain or fitted up for use. These they pay according to the largeness of the Provinces in which they dwell, the largest of which, they say, are een heel skatt, that is, they pay the full tribute; the lesser een half skatt, that is, half tribute; and so likewise for the rest. He that possesseth a Province of the whole tribute, pays two Patacoons, which they call Skattadaler, and others that have lesser possessions and half tribute, give one Patacoon; those which want mony, pay fish or skins, which are commonly of Foxes or Squirrils, of these 50, of the others one with a pair of Lapland shoes, are equal to a Patacoon: two pounds also of dry fish are of the same value; now to every pound of dried fish they allow five over, because so much is commonly lost in the drying. They call this pound with its addition Skattpund, that is the pound for tribute. They value their Rain-deers at 3 Dollars a piece, and pay the tenths of them, not each family, but every hundred. I have set the prices down here, because if any one had rather keep his Cattel, he can be forced to no more then after this rate. Now concerning the tenths they pay of skins, every housholder is taxed one white Foxe’s skin, or a pair of
- 85. Lapland shoes; if he hath neither of these, half a pound of dried Jack. This is the Tribute yearly received by the Crown of Sweden from Lapland, of which the greatest part is commonly by the Kings gracious favor allowed for the maintenance of their Priests, as was shewn in another place. Now because it is so far both by Sea and Land, before these commodities can be brought to the Kings Storehouses, besides the ordinary tax they give a pair of Lapland shoes, which they call Haxapalka, that is the price for carriage. This is all they pay to the King of Sweden, but besides they are tributary to the Crown of Danmark, and the great Duke of Moscovy, not as Subjects to these Princes, but upon the account of their receiving several advantages from their Dominions in their hunting and fishing. Those that are thus, are all the allotments of Torna beyond the mountains, who by reason of the liberty they have to bring down their Cattel from the mountains into the vallies in the Summer time, near the Sea shore, and taking the opportunity from thence of fishing, are taxed by the Danes, but not at above half the rate that they pay to the Swedes. These allotments are called Koutokeine, Aujouara, Teno, and Utziocki. The Laplanders also of the allotment of Enare in Kiemi, are in the same condition, who for fishing and hunting pay both to the Danes and Moscovites as well as to the Swedes: to the first one half, to the other a third part of what the Swedes receive. The tribute was in former time gathered when the Governor pleased, but afterwards only in the Winter, against which time it was all brought into Storehouses, each County having its proper place for that purpose. But when the place for their Markets and Fairs was determined, the Governor came thither and received it, which course they still take in this business. That this was also the time for receiving it, will appear from the account I shall give of their Fairs in the next Chapter.
- 86. CHAP. XIV. Of the Laplanders Fairs, and Customs in Trading. That we may not yet leave the Public concerns of the Laplanders, of which we have treated, let us proceed in the next place to consider their Fairs and common Markets, in which what Customs they anciently used is not so well known. Paulus Jovius saies that among the Laplanders he that had any thing to sell, after he had exposed his Wares, went his way and left them, and that the Chapman coming, and taking what was for his turn, left in the place the full value thereof in white furrs or skins. The reason why they did not speak and bargain with their Chapmen, he saies was, because they were a rustic People, extreamly fearful, and ready to run away from the very sight of a ship, or stranger. Others, that are of a more probable opinion, confess indeed that they used no words in their trading, but that it was not out of rusticity, want of cunning, or the like; but because they had a language quite different from others, and so peculiar to themselves, that they could neither understand, nor be understood of their neighbours: so that it was rather the barbarism, and roughness of their speech, then manners, that made them use this dumb way of traffiking. But of their language we shall treat in its proper place. Concerning their trading with their neighbours, it is most certain that it was performed without words, by nods and silent gestures: neither was it properly a buying and selling (for they did not of old use either gold or silver) but rather an exchange of one commodity for another. So that whereas Zieglerus tells us they did permutatione & pecunia commercia agere, we may justly doubt whether it be not
- 87. rather to be read nec pecunia, (unless happily he intend pecunia in the primary sense, and hath more respect to the original of the word, then to the acception now in use.) And truly this way of exchange among them, in those ancient times, was no less then necessary; when indeed, as well the neighbouring Countries, as the Laplanders were quite strangers to any current mony; and this we may understand from the Swedes, among whom there were in those daies either no coins at all; or else only such as had bin transported out of England and Scotland, the use of the Mint being then utterly unknown in that Country. And if at that time there was no mony in Swedland, it is certainly no great wonder there should be none in Lapland. But neither in after times, and when they were under the Jurisdiction of the Birkarli, could the Laplanders come to the use of mony; for they that were Lords over them, monopolizing the whole trade to themselves, did not give them mony for their commodities, but such other merchandise, as their Country stood in need of. In fine to this very day the Laplanders know no other mony but the Patacoon and half Patacoon; other coins whether of copper, silver, or gold, they do not so much value, which will give us to understand that the use of mony among them cannot be of any long date, for the Patacoon is but of later daies, and was never known before the discovery of the Mine in the Vale of Joachim. These Patacoons they value singly at 2 onces of silver a piece, whence it appears that as they had no other mony, so neither did this pass currant among them, but only by weight, and as if it were in the Mass: and I beleive was not at all in use, untill they were forced to pay tribute in that kind, of which I have discoursed before, and shewed that it was but of late instituted. But what Damianus means by his permutatione tantum annonam & pecuniam acquirunt, we cannot so easily guess; for we do not say that men barter and deal by exchange when mony is paid for a commodity: for to what end should those People seek after getting mony, which was in use neither among themselves nor their neighbours; so that perhaps here also we ought to read nec pecuniam, and then the sense runs, that they were not so sollicitous in getting mony, as in providing the
- 88. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com