The Journal of Supercomputing (2019) 75:2070–2105

https://doi.org/10.1007/s11227-018-2701-2

Fog computing: from architecture to edge computing

and big data processing

Simar Preet Singh1 · Anand Nayyar2 · Rajesh Kumar1 · Anju Sharma3

Published online: 23 November 2018

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Cloud computing plays a vital role in processing a large amount of data. However, with
the arrival of the Internet of Things, huge data are generated from these devices. Thus,
there is the need for bringing characteristics of cloud closer to the request generator,
so that processing of these huge data takes place at one-hop distance closer to that end
user. This led to the emergence of fog computing with the aim to provide storage and
computation at the edge of the network that reduces network traffic and overcomes
many cloud computing drawbacks. Fog computing technology helps to overcome
challenges of big data processing. The paper discusses the taxonomy of fog computing,
how this is different from cloud computing and edge computing technologies, its
applications, emerging key technologies (i.e., communication technologies and storage
technologies) and various challenges involved in fog technology.

Keywords Cloud computing · Fog computing · Fog network · Fog node ·

Architecture

B Simar Preet Singh

[email protected]
Anand Nayyar
[email protected]
Rajesh Kumar
[email protected]
Anju Sharma
[email protected]

1 Thapar Institute of Engineering and Technology, Patiala, India

2 Duy Tan University, Da Nang, Vietnam
3 MRSPTU, Bathinda, India

123
Fog computing: from architecture to edge computing and... 2071

1 Introduction to fog computing

The emergence of Internet of Things (IoT) technology leads to the development of

a wide variety of IoT devices. Smart home applications, tablets, smartphones, edge
routers, cellular base stations, smart traffic systems, smart energy meters, connected
vehicles, smart building controllers, etc. all use IoT devices. The rapid increase in
these varieties of network edge devices results in huge data transmission on a continual
fashion within a short span of time. These raw data need to be processed and respond
quickly.
Cloud computing consists of cloud data centers that are centrally deployed on a
global scale. Without fog computing technology, these data centers need to process
huge data coming from IoT devices. In addition, as the physical distance between the
cloud and the user increases, transmission latency and response time increase with
it. Users might become exhausted with the services provided [1–4]. The processing
speed in such a large environment is mostly dependent on the performance of the user
device. The approachable solution to these problems is the fog computing.
CISCO developed fog computing or fogging in 2014 [5]. This is also known as
fogging. Fog computing consists of multiple fog nodes arranged in a network which
are placed at just one hop from the user. Due to the placement of these multiple fog
nodes very close to the user, the data are processed locally inside fog nodes rather
than migrating the data to cloud server placed far away from the user [6,7]. Dast-
jerdi et al. [8] defined fog computing to be as the distributed computing environment
extending the services of cloud computing to one-hop distance from the user. Fog
computing technology provides networking, computation, storage and management
services between the end-users (edges) and the cloud data centers [9,10]. This supports
protocols for communication, mobility, resources to perform computing, distributed
analysis of data and integration of cloud for addressing latency-sensitive applications
that need minimum delay. In Marin-Tordera et al. [11], fog node is defined as a device
where the fog computing is deployed.
Figure 1 represents the transformation of cloud computing to the fog comput-
ing with adding functionalities with change of technology. Initially, users’ data were
directly processed by the cloud data centers. But with the emergence of fog computing,
latency-sensitive applications are processed by the fog nodes and latency in-sensitive
applications are processed directly by the cloud data centers.
Figure 2 represents fog nodes in a distributed arrangement within a fog network.
Fog network consists of multiple fog nodes, each communicating with one another
and also in sync with the cloud data center. Fog computing and cloud computing work
together to enhance quality and performance of any system.
In today’s era of computing, huge data are increasingly generated by the prolif-
erating IoT devices at the core network. These large quantity raw data are initially
processed by the fog networking or fog computing technology. Fog computing com-
prises of microclouds (fog nodes) that are placed at the edge of the data originator.
These microclouds have the mechanism how the data are to be processed before going
on to the network. Hence, fog computing structure is useful in big IoT data analytics.
This is on emerging phase and requires research initiatives on how effective knowledge
can be obtained for smart decisions [12].

123
2072 S. P. Singh et al.

Fig. 1 Transformation era from cloud computing to fog computing [12]

Fig. 2 Working of fog computing

123
Fog computing: from architecture to edge computing and... 2073

Continuously, huge data are generated by the IoT devices. According to the CISCO
estimation report, 847 zettabytes (ZB) of data will be generated by IoT devices by 2021
[13]. All these data are not useful and need to be mined. Fog computing overcomes this
processing of data at the edge of the network. As a result, fog computing overcomes
some of emerging big data problems [14].
Organization of the paper is as follows: This survey discusses fog computing tech-
nology with the introduction in the first part (Sect. 1) and related work in the second
part (Sect. 2). Section 1 explains the concept of fog computing. Section 2 discusses
the emergence of fog computing, its role in related technologies and how this tech-
nology is useful to the researchers in their areas. This section covers the description
about this technology stating how and in what all emerging fields this technology is
useful. Section 4 discusses the architecture of fog computing. Sections 5 and 6 discuss
the comparison of fog computing with different technologies, i.e., cloud computing
and edge computing. Section 7 explains key technologies in fog computing. Section 8
discusses the advantages of fog computing. Sections 9 and 10 present the applications
and challenges in fog computing.
Finally, Sect. 11 presents the conclusion and future scope of the paper.

2 Related work

Chiang and Zhang [15] discussed fog computing as a provider of various services at
the edge of the network making them closer to the user. They summarized new chal-
lenges and opportunities of fog computing and focused on how widely IoT network
can use fog computing platform and why there was a need for this new architecture.
They also discussed how various gaps in the technology can be filled with fog com-
puting architecture, which leads to increase in new business opportunities. Latency
constraints, network bandwidth constraints, resource constraints were some challenges
for fog computing cited in their work. Stojmenovic and Wen [6] discussed privacy and
security issues in fog computing. Study of man-in-the-middle attack was performed
considering the CPU and consumption of memory parameters. Dastjerdi et al. [8]
discussed the emergence of fog computing along with its key characteristics. They
presented a reference architecture and discussed the recent developments and appli-
cations. They categorized various research challenges like resource management and
energy minimization that were still in need for solutions in their work. They discussed
about the IoT, commercial products and the case study of the smart city as use case
for the Internet of Things.
An approach was discussed by Kukreja and Sharma [16] about the scalable dis-
tributed computing paradigms: fog, cloud and dew computing. They explained dew
computing as a novel structural layer in the available distributed computing hierar-
chy. This was placed as the base level for the fog computing and cloud computing
paradigms. From the cloud computing, the various divisions: hierarchical, vertical and
complementary to dew computing fulfilled the requirements of low and high-end com-
puting needs in day-to-day work and life. These novel computing paradigms decreased
the cost and enhanced the performance, specifically for applications and concepts, i.e.,
the Internet of Things (IoT) and Internet of Everything (IoE). More [17] defined fog

123
2074 S. P. Singh et al.

computing as a model in which various applications along with the data were stored
in devices which were placed at the edge nearer to the user rather than storing in the
existing cloud. They described fog computing as cloud computing extension along
with the various services of cloud computing that extends the computing to the user
edge. They discussed differences between cloud computing and the fog computing.
Tordera et al. [11] conveyed that fog computing was a “mini-cloud” on its first
attempt, which was using lots of edge node devices through wireless connections.
They surveyed and analyzed the role that edge devices play in the fog computing
definition, and later they summarized their lessons with the implementation. They
focused on core aspects of fog node and major research challenges toward fog nodes
in the future. They mentioned the location where these fog nodes are located. Table 1
describes their findings.
Mahmud and Buyya [29] discussed the fog computing environment along with
challenges in fog computing. They discussed five types of fog nodes, their configura-
tion and resource/service provisioning metrics. Security challenges in fog computing
were explained in their work. Saharan and Kumar [30] discussed the main characteris-
tics of this technology. They discussed the advantages along with various distinguish
parameters of the cloud computing and the fog computing. They analyzed its appli-
cations for IoT. Their approach conveyed that better QoS considering parameters as
delay and consumption of energy on the Internet, which is possible by this technology.
Yi et al. [31] discussed the unsolved cloud computing issues such as latency and loca-
tion awareness, etc. and suggested fog computing as a solution to these issues. They
discussed the fog computing definition and various issues that might originate during
design and implementation of fog technology and highlighted the issues related to
resource management, privacy, QoS, security and interfacing. Yi et al. [32] in another
paper discussed the various challenges in fog computing platform and presented several
application examples to explain the platform design. A prototype for fog computing
platform is implemented and evaluated for smart home applications.
Razouk et al. [33] discussed fog computing and proposed middleware architecture
from security point of view to solve the issues in cloud computing. Their architecture
acted as the smart gateway, which performed analytics at the edge of the network. Yet
another privacy and security issues were discussed by Alrawais et al. [34] and they
proposed techniques that helped fog computing to improve distribution of certificate
revocation information among the IoT devices. Using the proposed technique, their
model enhanced the security of entire system.
Roman et al. [35] discussed security aspects of edge computing. They shared various
research challenges in their work. Another work by Zhao et al. [36] focused Radio
Access Network (RAN) that exists between the traditional services and mobile edge
computing. Many researchers [37–40] discussed their work on edge computing.

Table 1 Fog node location [11]

Gateways [18–21]
Intermediate computing node [22–25]
Networking elements such as switches and [5,23–28]
routers

123
Fog computing: from architecture to edge computing and... 2075

Fog computing technology was used in almost every field including health care.
Rahmani et al. [41] presented an IoT-based remote health monitoring system that
implements prototype of Smart e-Health Gateway called UT-GATE and implemented
an IoT-based Early Warning Score (EWS) for health monitoring. They showed the
improvement in efficiency of their system using EWS. Vora et al. [42] proposed Fog
computing based patient monitoring system for Ambient Assisted Living (FAAL).
They also presented clustering algorithm which they used for data transmission that
helps in reduction in the communication infrastructure load.
Our work discusses fog computing and various new technologies that are emerging
with this technology. Applications, advantages and real- time usage of fog computing
are discussed. The paper also explains various challenges in the technology that leads
the way for researchers to work in those directions as a future research prospective.

3 Scope and motivation of work

Fog computing works by allowing some application processing to be performed at the

users’ end. The processing is performed with the fog server (fog node), which is placed
between the cloud and the user. This is mainly in a location physically closer to the user.
This allows some workloads of users devices be offloaded from the cloud at a location
that is closer to the user for processing. This speeds up applications processing time
that requires the low latency response. Thus, performance of the system is enhanced.

4 Fog computing architecture

Fog computing architecture is the arrangement of logical and physical software,

network elements and hardware for the implementation of useful network like IoT net-
work. The important architectural decisions include the positioning of the fog nodes,
their links capacities, topologies in which they are arranged, data bandwidth available
and their consumption. How the fog nodes are used and managed is also important
architectural decision.
Fog computing, together with the cloud computing, forms three layer architecture
[32,43] as described in Fig. 3:

Fig. 3 Basic architecture of fog

computing

123
2076 S. P. Singh et al.

i. Cloud (top most) layer: Cloud services and resources are contained in the top
most layer. It consists of the cloud server having huge storage capacity.
ii. Fog network (middle) layer: The middle layer consists of the fog network, com-
prising of the fog nodes connected to each other in the network. Fog nodes consist
of various networking devices like gateways, routers, switches.
iii. Things (bottom most) layer: The bottom most layer consists of the IoT devices
(things) consisting of sensors, smartphones, tablets and computers.
The bottom layer captures raw data in a continuous manner from the user end and
simultaneously transfers these data to the edge node. These raw data are reached at the
edge node which is closest to the user. These raw data are processed, and computation
on these raw data is performed at the fog node. These fog nodes are placed in the
middle layer. Analytics is performed at the fog computing network with the help of
active fog nodes. This analytics is performed near to the user, and response is sent for
further processing immediately, thus reducing latency in transmission, and provides
quick response. The results after computation and corresponding action are updated
to the cloud on the top most layer. Thus, fog computing helps to reduce the packets
transmitted on the network between the fog and the cloud. This helps in improving
bandwidth, better performance and faster transmission of packets over the network.
Marin-Tordera et al. [11] conveyed this architecture as a three fog computing lay-
ers for analysis of big data in smart cities. According to them, the first fog layer is
the intermediate computing node that comprises of computers having efficient com-
puting power. The second layer is termed as edge computing nodes that contain tiny
computing nodes like the mobile phones. The last and the third layer is called as the
lowest fog layer that comprises of sensors that have sensing capabilities. Taneja and
Davy [44] described the basic architecture as the tier division. Their tier division is
shown in Fig. 4. They described the fog computing architecture using tier division as
represented in Fig. 5.

5 Differentiating cloud and fog computing: analysis

The cloud computing has large data centers each having an infinite storage space that
can be used to store large amount of data [45]. Data follow the path on the Internet to
reach to the cloud server as cloud services are provided by the third party vendors to the
end-users. Thus, migrating data to the cloud server and then receiving the response and
vice versa consume much amount of time; especially, this is important when the case of
real-time applications is considered, where the response is required within a fraction of
a second. Cloud computing takes an interminable time to response as the request has to

Fig. 4 Tier division [44]

123
Fog computing: from architecture to edge computing and... 2077

Fig. 5 Architecture of fog computing

cover through a large network of connecting devices like routers, gateways, repeaters.
This shortcoming of cloud computing is overcome by fog computing [46–50].
Figure 6 represents the fog computing in comparison with cloud computing. Fog
computing is cloud placed near to the user, which has a low computational capacity and
small storage space. Fog network is composed of fog devices like routers, gateways,
repeaters [51–53]. All the request from the end-users come to these fog devices initially.
Computations are performed on the fog devices and are communicated with request–
response cycle. The requests which do not require a rapid response can be sent to
cloud for handling. Hence, fog computing and cloud computing are two different
technologies, but they work together to provide greater response time, higher efficiency
and better performance. The comparing parameters between the cloud computing and
the fog computing considering various characteristics are described in Table 2.

6 Differentiating edge and fog computing: analysis

Edge computing pushes localized processing in the advanced manner, i.e., closer to
the data source. For attaining this, the connection between Programmable Automation
Controllers (PAC) and the sensors is made. These PACs handle all the processing and
communications. Table 3 shows the characteristics of fog and edge computing.

123
2078 S. P. Singh et al.

Fig. 6 Fog computing and cloud computing

Table 2 Difference between cloud computing and fog computing [17,30,54]

Characteristics Fog computing Cloud computing

Latency Low High

Delay Very low High
Server nodes location Edge of network (local) In-between the internet
Security Can be defined Undefined
Client and server distance One hop Multiple hops
Location awareness Yes No
Prone to attacks on data in transmission Very less probability High probability
Geographical distribution Distributed Centralized
Mobility Supports mobility Limited
Server nodes count Very large Few
Interactions in real time Supports Supports

Table 4 shows difference between the fog computing and edge computing. The
venn diagram showing the comparison between these technologies is shown in Fig. 7.

7 Key technologies

Various key technologies in fog computing involve [55,59].

7.1 Communication technologies

Various communication technologies have emerged with fog computing like Software-
Defined Network (SDN) [60,61], Network Function Virtualization (NFV) [62],

123
Fog computing: from architecture to edge computing and... 2079

Table 3 Fog and edge computing characteristics [55,56]

Characteristics Fog computing Edge computing

Hosting application Yes Less scope

Provides data service at edge Yes Yes
Management of device and Yes Yes
application
Security E2F, data protection, session Point to point partial
and hardware level solution VPN, FW
Resource pooling/elastics Yes No
compute
Interchangeable hardware Yes No
Windows support Yes TBD
virtualization
Real-time high availability Yes No
control

Table 4 Difference between fog computing and edge computing [57,58]

S. no. Fog computing Edge computing

1 Device independent, intelligent, and Device and services aware, no

knowledge of whole fog network knowledge of entire network
2 Controls all devices in the network Limited control in the edge network
3 Fog computing extends cloud to fog level in Edge computing is cloud unaware
a continuum
4 Complete network scope Limited network scope
5 Enables multiple IoT verticals and provides No IoT vertical awareness
support for them
6 Supports integration of multiple verticals No IoT vertical integration
7 Versatile fog nodes that perform variety of Edge device is controlled and
tasks like website hosting and management communicated with edge
controllers
8 End-to-end security Security scope is limited to devices

concept of 5G technology [63,64], Content Distribution Network(CDN) [65–67],

Long-Reach Passive Optical Network (LRPON) [68–72].
I. Software-defined network (SDN) Scalability, flexibility, ability to program and
global knowledge are provided by SDN architecture [73]. SDN separates data
and control layers [74]. From the technical point of view, SDN is a new network
architecture which comprises of three different layers as shown in Fig. 8. The third
layer, i.e., application layer, is for covered applications, management applications
and business applications, and places the demands for the network onto the control
layer.
(a) Components of SDN SDN works on manageable networking devices. These
devices consist in two parts: data plane and control plane. Earlier networking

123
2080 S. P. Singh et al.

Fig. 7 Edge computing, fog computing and cloud computing

Fig. 8 SDN network architecture

devices were tightly coupled that means the data plane and control plane can-
not be available separately. In tightly coupled devices, modifications are not
possible as both the data plane and control plane are coupled with each other.
To overcome above limitation, SDN technology gives ability to purchase data
plane and control plane separately. Network intelligence is centralized in one
network component by the SDN. This leads to the disassociation of forwarding
process of network packets (i.e., data plane) from the routing process (i.e., con-
trol plane). It decouples data plane & control plane, by which we can design
our own application/code as per our requirement, and we make that code com-

123
Fog computing: from architecture to edge computing and... 2081

patible with control plane and data plane. Now only code decides, and device
will work like a switch, like a router or a firewall (security device) [60].
i. Data plane Data plane in networking device works like forwarding plane,
which actually forwards the data; for example, physical ports on switch
work as data plane. Data plane is also known as forwarding device, open
flow switch and SDN switch. There is no intelligence in data plane. It is
like dummy device.
ii. Control plane Control plane in networking device works like controller
that actually controls the flow of data. It is like IOS (Internetwork Oper-
ating System) of device which runs from CLI (Command Line Interface)
mode. Control plane is also known as SDN controller.
Figure 9 shows the components of SDN. This comprises of data plane at the
bottom, followed by the control plane and then the application plane. All the
application logics reside on the application plane. Control plane consists of
the SDN controller that controls all the flows of the network that is coming in
and moving out of the SDN. The bottom part, i.e., the data plane, contains the
network elements that perform the processing functions.
II. Network function virtualization (NFV) NFV is an emerging technology that can
be used in 5G technology. It is the subset of Software-Defined Networking (SDN).
NFV decouples hardware from software. The network operators are populated
with the large hardware appliances. They require large hardware equipment for
the number of users so that they can meet their growing demands and new services.
In order to launch the new services, we require another variety of hardware, space
and power. It is very difficult to achieve because we need increase cost of energy, a

Fig. 9 SDN components

123
2082 S. P. Singh et al.

lot of investments and complex hardware designs. NFV is aimed to remove these
problems while using IT virtualization technology to consolidate many hardware
equipment types like server, switches and storage that are located in data center.
Thus, the aim of NFV is to make network architecture simpler and faster.
NFV consists of various components. NFV Infrastructure (NFVI) provides the
infrastructure to the NFV. This contains virtualization layer on which various
modules will be deployed. These modules consist of virtual computation, virtual
storage services and virtual network resources. This virtualization layer is hosted
above the physical hardware resources. The administrators can access and manage
this all using the NFV management console. On the top of the NFV Infrastructure
(NFVI), there lies the Virtualized Network Functions (VNFs). This comprises of
multiple VNFs deployed over the NFV Infrastructure (NFVI). Figures 10 and 11
show the components of NFV.
(a) Benefits of NFV The benefits of NFV are as follows.
i. It provides more flexibility as it separates the hardware and software.
ii. It is much more scalable and elastic.
iii. The network function can be localized.
iv. It will reduce equipment cost and consumption.
v. It will increase the speed of time to market by minimizing operator cycle
of innovation.
vi. Operators can share resources that mean resources can be utilized more
effectively.
vii. Lower risks.
III. Concepts of 5G technology 5G wireless technology is the 5th generation of
wireless technology. It provides wireless communication with almost no limi-
tations. Every generation has some limitations like in first generation; it was a

Fig. 10 NFV architecture

123
Fog computing: from architecture to edge computing and... 2083

Fig. 11 Detailed components of NFV

Fig. 12 Fifth-generation (5G) technology

speed limitation, call limitation and audio/video limitations. As the years pass on,
with the progress in the technology, 5th generation has up to no limitations. This
generation can be called a real wireless world. While sitting at the one end of the
world, we can communicate with the person sitting at the other end of the world
without any interruption, without any speed limitation or communication limita-
tion and with incredible transmission speed [75]. Now we can transfer huge files
up to GBs per second. This concept is only theory and has not launched in real
world so far. The concepts involved in 5G technology are represented in Fig. 12.

123
2084 S. P. Singh et al.

(a) 5G technology opportunities in context with fog computing are described

below.
i. It will offer worldwide cellular phones.
ii. It will provide extraordinary data capability.
iii. It will provide high connectivity. For example, if we are in some other
continent and want to communicate with your friend residing in another
continent, we can easily contact with high connectivity even in crowded
areas.
iv. It allows transfer of data at a very high rate. (100 times more than the
typical data rate even for high mobility.)
v. It will provide more power and features in handheld phones. We can roam
and do your work while communicating.
vi. It will offer large phone memory. Previously phone memory was up to 1
GB or 2 GB, but now it can be up to 120 GB.
vii. It will provide more dialing speed. We can connect a person within sec-
onds.
viii. It will provide more clarity in audio and video.
Figure 12 shows about the advantages that fifth generation offers.
(b) Basic architecture of 5G technology
Figure 13 shows the layered architecture of both the normal ISO/OSI model
and 5G technology, and Fig. 14 describes the basic architecture of 5G technol-
ogy. The Open Wireless Architecture of 5G layer is the combination of both
the physical layer and data link layer of OSI Model. The upper network layer
and the lower network layer of 5G technology architecture are a combination
of network layer of OSI Model. Similarly, the Open Transfer Protocol of 5G
architecture is a combination of session layer and transport layer and applica-
tion (service) is the combination of application layer and presentation layer of
OSI Model.

Fig. 13 Layered architecture of ISO/OSI model and 5G technology

123
Fog computing: from architecture to edge computing and... 2085

Fig. 14 Basic architecture of 5G

technology [76]

IV. Content distribution network (CDN) Content Delivery Network or Content Dis-
tribution Network, simply put, is a collection of thousands of thousands computes
(fog devices). This is simply the network of attached computers (fog network).
All of these computers are ready to serve any request sent to them. They deliver
requested Web pages and other Web content to the user. The goal of CDN is
to provide high availability and high performance to users. CDNs provide faster
request–response cycle in case of Web pages and thus give better performance dur-
ing high traffic. The speed of delivery of contents is based on users geographical
location. The closer the CDN server is to the user geographically, the faster the user
will get the requested contents. Figure 15 shows the Content Delivery Network.

(a) Why we need CDN

To understand the need of CDN, we must be knowing how we can construct
a typical Web page [65]. A client that could be a Web browser, mobile device
or any other computer connected to the Internet sends a request to the server.
For example, when one type in his browser, www.google.com, that browser
(client) sends request to google.com (server) and asks for the Web pages. Then,
the Google server does some computations and sends raw html page back to
the client. This html page has multiple links like CSS, JavaScript and images.
Figure 16 shows this scenario. This same process of request and response
cycles works in case of fog computing also while communicating with the fog
devices.
After client downloads the initial html page, it makes another html requests
to get CSS, JavaScript and images. Thus, client sends so many requests for
just one page www.google.com. The CSS, JavaScript and images are static

123
2086 S. P. Singh et al.

Fig. 15 Content delivery network (CDN)

Fig. 16 Layered architecture of OSI/ISO model and 5G technology

contents, i.e., contents that do not change. They remain static with time rather
than dynamic contents [67].

In a small company, a center server serves all the requests from its clients.
The same server manages many connections from a single client. This leads
to shortage of bandwidth or capability to serve large number of requests from
millions of clients even you may have thousands of servers.
Content Distributed Network (CDN) provides the better solution for such prob-
lems. CDN helps to remove all of that traffic, and it also takes over the load

123
Fog computing: from architecture to edge computing and... 2087

Fig. 17 Layered architecture of ISO/OSI model and 5G technology

of all the CSS, images and JavaScript (js) that the original server was not able
to serve and also frees up a lot of bandwidth to serve in future and for more
dynamic stuff [66]. This scenario is shown in Fig. 17.
Thus, the www.google.com or any Internet skilled companies like Facebook,
Linkedin serve dynamic contents and links to CSS, images and JavaScript are
redirected to CDN provider. Thus, they uses a CDN to serve their needs of the
static contents and their servers are free to serve all the extra requests.
(b) Advantages of using CDN
i. As CDN is a collection of computer spread across the earth, it stores a
cached version of its content in multiple geographical locations. So the
client from Brazil, for example, would get the closest network of comput-
ers and would get all the static contents from the closest location of that
CDN machine rather than traveling the full width of US server and back.
Thus, it provides faster content delivery to end-users.
ii. It provides a lot of free bandwidth to the server.
iii. One of the major advantages is warm cache. If there is a new feature com-
ing out and all the browsers do not have those new static contents, CDN
helps to warm the cache earlier when we are testing the feature, and all
the static contents will be cached. So all of those resources are available
for the client for the new features.
iv. It provides security of data in case of Denial-of-Service (DoS) attacks.

V. Long-reach passive optical network (LRPON) Emergence of technology leads

to the increase in span of Passive Optical Network (PON) broadband access to
a larger extent. This extended reach Passive Optical Network (PON) is called
as Long-Reach Passive Optical Network (LRPON). Generally, this extension is

123
2088 S. P. Singh et al.

Fig. 18 Long-reach passive optical network (LRPON) [80,81]

from 20 km (in case of PON) to 100 km or higher (in case of LRPON). This
technology enables broadband access among large number of users with reduced
cost [77–79]. LRPON network is shown in Fig. 18.
LRPON is designed for long-range coverage of broadband access. This helps to
serve large number of users. Revenue and data loss occur for network operator
and customers if any network failure occurs. The architecture comprises of fiber
cable that connects customer directly with the local exchange and has an optical
splitter which connects users to the shared fiber [82]. This technology is very
useful for voice over IP (VoIP) and IP video delivery. Allocation of resources,
management, service aggregation and control are the various research challenges
in this technology [83,84].
Advantages of Long-Reach Passive Optical Network (LRPON) are as follows
[68,69,71].
(a) It provides large split 512–1024 instead of 32–64 as shown in Fig. 18.
(b) It provides 10 Gb/s peak to customers (∼ 10,000 times DSL rate) and 10–
20 Mb/s sustained bandwidth (∼200 times DSL rates).
(c) Future bandwidth can be ∼ 100 times greater via WDM.
(d) 5600 local exchanges in UK can be reduced to ∼ 100.
(e) Optical networks are more reliable than copper.

7.2 Storage technologies

Storage is the key technology in fog computing. Bringing storage closer to the user
enhances security and privacy and enhances the low latency. Fog network comprises
of heterogeneous nodes that have small computing capabilities, but vary in physical
components. Thus, how to deploy the applications on the fog devices and in a proper
manner is the major concern [58,85,86].

8 Advantages of fog computing

Various advantages of fog computing [30,87–89] include:

123
Fog computing: from architecture to edge computing and... 2089

I. Efficiency and cost It reduces the need for bandwidth using distributed strategy
which results in lower costs and improves efficiencies. Fog computing is placed
at the edge of the network, which is one hop from the user. This is the place
where all the computations are carried out [90,91]. This reduces bandwidth and
hence enhances efficiency.
II. Location awareness Using fog computing, latitude and longitude values of the
fog devices can be captured. This provides location awareness and minimum
latency and improves the Quality of Service (QoS) in real-time interactions as
this is placed very close to the user [79,92].
III. Geographical distribution It provides widespread geographical distribution
which consists of a network of fog nodes that are geographically distributed and
are connected to the centralized cloud server [93,94].
IV. Mobility: It provides mobility and closeness as it is operating at the edge of the
network. This is simply like we are having a moving VM that keeps on moving
as we are navigating from here and there and in that VM, and all the data of the
user are stored [95,96].
V. Enhanced security Fog computing provides enhanced security. Policies and
procedures can be made to protect the fog nodes in the network [97,98].
VI. Deeper privacy control Fog computing provides greater privacy over cloud
computing. The sensitive information is analyzed locally near to the user, rather
than sending the sensitive information on the network to the cloud. Fog devices
can easily track and managed by the IT team within the organization, hence
enhancing the privacy [99,100].
VII. Better interconnectivity As the fog network is nearer to the user, connectivity
capacity is good between the user and the fog node. This helps to get faster
response and better analytics using the fog nodes [101].
VIII. Reduced latency The analysis time gets reduced with the emergence of fog
computing. This is because of bringing computation closer to the user. Since,
response time of computation is quick as compared to cloud computing, hence
reducing the latency in computation [102,103].
IX. Better quality of service (QoS) Quality of Service (QoS) depends upon many
factors like reliability, link health, bandwidth of the link. All these factors per-
formed well in fog computing as compared to cloud computing [104–106].
Thus, Quality of Service (QoS) is enhanced using fog computing technology.

9 Applications

Fog computing, being placed closer to the user, is used in many applications, especially
those that are latency-sensitive. Few applications are as follows:
I. Smart grid Fog computing plays a vital role in energy management in smart grid
application. Comparing the energy demand, these devices have the capability
of switching to the alternative energies like wind and solar energy system [107–
109]. In this application, grid devices like sensors, which are present at the
edges, are attached to the fog devices. These grid devices generate the data and

123
2090 S. P. Singh et al.

Fig. 19 Smart grid

Fig. 20 Streetline parking automation system [22]

fog devices issues commands to the actuators for managing the generated data
[6,24,110]. Figure 19 represents smart grid application.
II. Streetline parking automation system The streetline parking automation sys-
tem [11,22] is represented in Fig. 20. The vehicles like cars have the sensors
attached to them. This is represented as embedded systems and sensor layer in
Fig. 20. These sensors remain continuous in connection with the fog network.
The sensors emit continuous signals/data that are received by the nearest fog
node inside the fog network. The fog network forms the multi-service edge
layer that consists of the fog nodes connected together. All the intelligence is
performed here. This layer is in link with the cloud layer using the Multi-Protocol
Label Switching (MPLS) and other protocols.

123
Fog computing: from architecture to edge computing and... 2091

Fig. 21 Application: intelligent transportation system [22]

III. Intelligent transportation system Another application of fog computing is

the Intelligent Transportation System. In this, all the vehicles are having sensors
that are connected with the fog devices. This also provides an ease for vehicle-
to-vehicle communication. Figure 21 describes the Intelligent Transportation
System. This system involves many communications including satellite com-
munications, vehicle-to-vehicle communications, safety systems, traffic signs,
fleet management [98,111].
IV. Smart health-care monitoring system Health monitoring or observations,
which have significant variations, are the focus of applications during analytics
of data in biomedical area for correctly identifying smart health-care decisions
in the future [12]. In real time, this is possible by three-level hierarchal structure
comprising of fog nodes rather than using cloud computing technology inde-
pendently.
U-Fall, a real-time monitoring application proposed by Cao et al. [112], has
three main sections: front end, back end and communication module. In this
application, front end and back end both take independent decisions, and then,
their collaborative results form the actual solution. This independently detec-
tion technique reduces the wrong alarm rate of the application and improves the
accuracy of the collaborative result. Gia et al. [113] utilized fog computing as
a smart gateway for computing various complicated methods. Aazam and Huh
[114] proposed Emergency Help Alert Mobile Cloud, a smart phone application
that uses fog nodes for preprocessing the data and offloading the response to

123
2092 S. P. Singh et al.

Fig. 22 Smart health-care monitoring system

appropriate department (like ambulance) having contact details already stored

within the application. They proposed the application for real-time use. Figure
22 describes smart health-care monitoring system.
V. Real-time video analytics Fog computing provides well-organized platform for
the applications that contains video content. This technology overcomes latency
issues in the packets transmitted over the network, thus making the video qual-
ity rich to the user [12,115]. This technology plays a vital role in real-time data
analysis. Hong et al. [116] performed face recognition process by identifying
data collection process at various cloud instances. They used three-level hierar-
chal structure comprising of fog nodes to act upon smart cameras using motion
detectors. Fog computing is also used in on-demand video streaming. Zhu et
al. [117] performed real-time analytics on video data for surveillance cameras
and enhanced the communication in a virtual desktop structured system. Fig-
ure 23 represents video analytics.
VI. Smart traffic lights system Smart traffic lights system is used for monitoring
instances like accidents, flow of the traffic and collects the related data for future
analysis [118,119]. In smart traffic lights system, many sensors interact with
this system for obtaining the information [12]. Information collected using this
system further helps in prevention of accidents by generating warning signals in
connected vehicles. Stojmenovic and Wen [120] proposed a system with video
surveillance cameras for sensing and monitoring the accidents and sending alert
alarm to an ambulance. Further, this system has emergency traffic light support
to pass the ambulance in emergency situations. Many researchers [92,121–129]
also presented their work on smart traffic lights system using fog computing
technology. Figure 24 represents smart traffic lights system.
VII. Gaming The emergence of cloud computing allowed gaming users to play with
more than one players. With the emergence of fog computing technology, well-

123
Fog computing: from architecture to edge computing and... 2093

Fig. 23 Video analytics

Fig. 24 Smart traffic lights system

organized platform is introduced for online gaming users. With this, essential
gaming hardware along with multiple players for playing is made available
[12]. This results in rapid growth of online gaming players. Wang and Dey
[130] discussed mobile gaming application framework that works with cloud

123
2094 S. P. Singh et al.

Table 5 Application of fog computing in various fields [30,89,120,136]

Field Description

Health maintenance Real-time monitoring system for patients in critical care units
[137–139]
Gas and oil Parallel monitoring system for leakage, theft, fire etc.
Agriculture Smart farms with crop monitoring and irrigation control systems
Management of energy Control of smart grid having privilege to switch between various
sources of energy
Travel management Vehicle health monitoring systems for buses, trucks
Retail management Shopping cart tracking and automated billing systems
Smart home Ease and security systems like weather control, fire alarms, intruder
detection
Smart traffic lights Presence of pedestrians and vehicles can be detected using smart street
lights

computing. They discussed quality issues with high latency in the application.
Zhao et al. [131] targeted Quality of Service (QoS) in fog computing to obtain
high gaming Quality of Experience (QoE).
VIII. Mobile big data analytics Emerging area in research for big data architecture
in mobile and in cloud computing nowadays is the big data processing [132–
134]. Fog computing overcomes cloud computing issues by providing various
resources for large-scale data processing. Thus, big data processing is handled
by integration of fog and cloud technologies. This reduces total computation
power in processing big data. This forms the key technique to perform analytics
on large generated data in context of IoT [135].
There are various applications of fog computing in different fields as described in
Table 5.

10 Challenges in fog computing

In spite of fog computing that brings computing capabilities closer to the user and
provides many additional benefits over cloud computing, still fog computing, in its
initial face, is facing many challenges. The various challenges in fog computing are
[140–142]:
I. Load balancing is a concept in fog computing which is very useful to achieve an
energy efficient system. The research work proposed by Varghese et al. [143]
contains heterogeneous nodes that range from sensors to user devices and to
routers, switches and mobile base stations. To perform general purpose com-
puting on these different types of resources, both horizontal scaling and vertical
scaling are required. The authors have not considered the vertical scaling. Both
the horizontal scaling and vertical scaling can be considered as the resources
which might not be under the ownership of same server and might not be so
tightly coupled, as the servers in a data center [79,107,144,145].

123
Fog computing: from architecture to edge computing and... 2095

II. Deployment strategy The biggest challenge is the deployment strategy. How the
work load is deployed on the fog nodes placed in the fog network? [90,146,147].
III. Connection policy Another challenge is how the fog nodes are connected with
each other and how to use the available fog nodes for handling the work load
[148].
IV. Resource provisioning may need to be carefully designed by using the priority
model for better scheduling of resources [31,32]. Resource discovery and shar-
ing are critical for performance of applications in fog computing. The resource
sharing optimization techniques can be formulated for maximizing the avail-
ability of resources and minimizing the energy consumption [8,31,32,149]. VM
scheduling needs a new design to provide an optimal solution for scheduling
VMs [32,106,150,151].
V. Heterogeneous environment Since fog nodes comprise of different devices like
mobile phones, sensor, desktop, laptops, thus heterogeneity is the challenge in
fog computing, i.e., how different devices communicate each other for handing
the tasks? [152–155].
VI. Energy minimization Dastjerdi et al. [8] considered fog computing as a network
consisting of a large number of fog nodes within which the computation is
distributed. This can be less energy efficient as compared to cloud computation
model. The improved model can be proposed by taking energy minimization
into consideration. Many researchers [145,152,156] have focused on energy
minimization in fog computing.
VII. Tasks offloading Task offloading is the process of executing the tasks on the fog
nodes and on the cloud. This selection of execution depends upon various param-
eters like tasks that require high computing capabilities which are offloaded to
the cloud for execution and tasks that require less computing capabilities which
are handled on the fog nodes. Task offloading can also be carried out considering
the latency like tasks that are latency-sensitive are executed on fog nodes and
non-latency-sensitive tasks that are offloaded to the cloud [157–160].
VIII. Quality of service (QoS) Fog computing challenge is to increase the QoS. This
includes the link quality, energy aspects, bandwidth issues, traffic constraints
on the network parameters that affect the quality of the link. The stronger the
quality of the link, the best will be the quality of service [31,32,54].
IX. Fog node security and privacy Security is an important challenge. Data follow
from the user to the fog network and to the cloud. All computations are car-
ried out at fog nodes and in the fog network; thus, securing the fog network
is important. Security and privacy need to be examined at every stage of fog
computing paradigm. Malicious user might pose as a device and may access
information without authorization leading to security violation [8,15,161]. To
mitigate the risk, encrypted data exchanged between the devices and the fog
controller can be proposed using Public Key Infrastructure. In addition, mon-
itoring of the data patterns received by the fog controller may be performed
to detect the fake devices signatures for taking necessary action to block it off
the network access control. Security also comprises secure data computation,
secure data storage, access control, network security etc. [122,162–167]. Pri-

123
2096 S. P. Singh et al.

vacy comprises of data privacy at end devices, intrusion detection, rogue node
detection, etc. [162,166–176].
X. Software-Defined Network handles issues due to the reliability of wireless links
and mobility of devices in the network and maintains connectivity. Issues like
fog controller placement, communication between different controllers like con-
stantly connected controller (at edge infrastructures) or intermittently connected
controller (at end devices) and the designing of the distributed SDN system to
meet the fog computing requirements can be considered [15,31,62].
XI. The authors of [31,32] proposed Network Function Virtualization technique that
can create virtual machine instances on the fog controller to handle functions
like gateway, intrusion detection, firewall, etc as per requirement and help in
configuration of the network under dynamic conditions. The performance of
virtualized network appliances needs to be addressed, which has two aspects:
latency or throughput of virtualized network appliances and other is how to
achieve efficient placement, instantiation and migration of virtual appliances in
a dynamic network [62].

11 Conclusion and future scope

Fog computing brings computing nearer to the user. Fog computing is not a different
technology. It works in coordination with the cloud computing. This helps in over-
come drawbacks of cloud computing. Fog computing is useful in applications that
are latency-sensitive and require computations to be performed near to the user. This
technology reduces congestion on the network, hence improving the overall network
performance.
This work discusses the fog computing covering the taxonomy of this technology’s
architecture, which involves with cloud computing, to reduce the computation time.
The analysis and comparison of different technologies (like edge computing, fog
computing and the cloud computing) are discussed. This work covers the drawbacks
of cloud computing that overcomes with fog computing technology.
With the emergence of IoT and widely increasing trend of IoT devices, fog comput-
ing technology has become the need of the hour. Fog computing performs computation
at the edge of the network, closer to the user. This work covers the key technologies
that have originated with fog computing.
This paper discusses various research challenges in fog computing. Being an
emerging technology, this technology has challenges in resources distribution, tasks
offloading, performance and capacity of the network links, etc.
In future work, these challenges may be worked on and many new advancements
in the area of fog computing may be originated. In spite of many opportunities of
fog computing for cost reduction, proficiency and business continuity, there are many
challenges also that are being faced by industrial data management. SDN and NVF
both when used together lead to cost reduction and increased network scalability.
With the emergence of IoT devices, huge load is moving on the network. This leads
to arising big data problems. Fog computing technology is the need to refine this load

123
Fog computing: from architecture to edge computing and... 2097

on the network. This technology efficiently regains network performance in terms of

bandwidth consumption, data storage, processing of data and data movement.
Fog computing is necessary for the emerging technologies like IoT, 5G, virtual
reality. Remote operations and data incentive are also supported by fog computing
technology. Fog computing manages the bandwidth, energy, and computing needs.
This is also resilient to network disruption and adapts to bandwidth changes. Fog
computing enables trusted, rapid and secure transmissions over the network.

References
1. Enokido T, Doulikun D, Takizawa M (2017) An energy-aware load balancing algorithm to perform
computation type application processes in a cluster of servers. Int J Web Grid Serv 13(2):145. https://
doi.org/10.1504/IJWGS.2017.10004125. URL http://www.inderscience.com/link.php?id=10004125
2. Liu Z, Li J, Wang Y, Li X, Chen S (2017) HGL: a hybrid global-local load balancing rout-
ing scheme for the Internet of Things through satellite networks. Int J Distrib Sens Netw
13(3):155014771769258. https://doi.org/10.1177/1550147717692586. URL http://journals.sagepub.
com/doi/10.1177/1550147717692586
3. Muck TR, Ghaderi Z, Dutt ND, Bozorgzadeh E (2017) Exploiting heterogeneity for aging-aware load
balancing in mobile platforms. IEEE Trans Multiscale Comput Syst 3(1):25–35. https://doi.org/10.
1109/TMSCS.2016.2627541. URL http://ieeexplore.ieee.org/document/7740903/
4. Jiang F, Liu Y, Wang B, Wang X (2017) A relay-aided device-to-device-based load balancing scheme
for multitier heterogeneous networks. IEEE Internet Things J 4(5):1537–1551. https://doi.org/10.
1109/JIOT.2017.2677975. URL http://ieeexplore.ieee.org/document/7870597/
5. Cisco Fog Computing Solutions: Unleash the Power of the Internet of Things (2015) URL https://
www.cisco.com/c/dam/en_us/solutions/trends/iot/docs/computing-solutions.pdf
6. Stojmenovic I, Wen S (2014) The fog computing paradigm: scenarios and security issues. In: 2014
Federated conference on computer science and information systems (FedCSIS), pp 1–8. https://doi.
org/10.15439/2014F503. URL https://fedcsis.org/proceedings/2014/drp/503.html
7. Rani S, Ahmed SH (2018) Secure edge computing: an architectural approach and industrial use case.
Internet Technol Lett 1:e68
8. Dastjerdi AV, Buyya R (2016) Fog computing: helping the internet of things realize its poten-
tial. Computer 49(8):112–116. https://doi.org/10.1109/MC.2016.245. URL http://ieeexplore.ieee.
org/document/7543455/
9. Deshmukh UA, More SA (2016) Fog computing: a new approach in the world of cloud computing.
Instr Technol 49
10. Gohar M, Ahmed SH, Khan M, Guizani N, Ahmed A, Rahman AU (2018) A big data analytics
architecture for the internet of small things. IEEE Commun Mag 56(2):128–133
11. Marn-Tordera E, Masip-Bruin X, Garca-Almiana J, Jukan A, Ren G-J, Zhu J (2017) Do we all
really know what a fog node is? Current trends towards an open definition. Comput Commun
109:117–130. https://doi.org/10.1016/j.comcom.2017.05.013. URL http://linkinghub.elsevier.com/
retrieve/pii/S0140366416307113
12. Anawar MR, Wang S, Azam Zia M, Jadoon AK, Akram U, Raza S (2018) Fog computing: an overview
of big Iot data analytics. Wireless Commun Mob Comput. https://doi.org/10.1155/2018/7157192
13. Cisco estimation report. URL https://www.cisco.com/c/en/us/solutions/collateral/service-provider/
global-cloud-index-gci/white-paper-c11-738085.html#_Toc503317525
14. Hussain F, Alkarkhi A (2017) Big data and fog computing. In: Internet of Things, pp 27–44. https://
doi.org/10.1007/978-3-319-55405-1_3
15. Chiang M, Zhang T (2016) Fog and IoT: an overview of research opportunities. IEEE Internet
Things J 3(6):854–864. https://doi.org/10.1109/JIOT.2016.2584538. URL http://ieeexplore.ieee.org/
document/7498684/
16. Kukreja P, Sharma DD (2016) A detail review on cloud. Fog Dew Comput 5(5):9
17. More P (2015) Review of implementing fog computing. Int J Res Eng Technol 4(06):335–338
18. Rahmani A-M, Thanigaivelan NK, Gia TN, Granados J, Negash B, Liljeberg P, Tenhunen H (2015)
Smart e-health gateway: bringing intelligence to Internet-of-Things based ubiquitous healthcare sys-

123
2098 S. P. Singh et al.

tems. In: IEEE, pp 826–834. https://doi.org/10.1109/CCNC.2015.7158084. URL http://ieeexplore.

ieee.org/document/7158084/
19. Aazam M, Huh E-N (2014) Fog computing and smart gateway based communication for Cloud of
Things. In: IEEE, pp 464–470. https://doi.org/10.1109/FiCloud.2014.83. URL http://ieeexplore.ieee.
org/lpdocs/epic03/wrapper.htm?arnumber=6984239
20. Aazam M, Huh E-N (2015) Fog computing micro datacenter based dynamic resource estimation and
pricing model for IoT. In: IEEE, pp 687–694. https://doi.org/10.1109/AINA.2015.254. URL http://
ieeexplore.ieee.org/document/7098039/
21. Gia TN, Jiang M, Rahmani A-M, Westerlund T, Liljeberg P, Tenhunen H (2015) Fog com-
puting in healthcare Internet of Things: a case study on ECG feature extraction. In: IEEE, pp
356–363. https://doi.org/10.1109/CIT/IUCC/DASC/PICOM.2015.51. URL http://ieeexplore.ieee.
org/document/7363093/
22. ASE International Conference on Big Data (2015) Academy of Science and Engineering, Association
for Computing Machinery. In: ASE international conference on social informatics, a hierarchical
distributed fog computing architecture for big data analysis in smart cities, 00000 OCLC: 956994157.
URL http://dl.acm.org/citation.cfm?id=2818869
23. Bonomi F (2011) The smart and connected vehicle and the Internet of Things, enabling technologies.
URL http://tf.nist.gov/seminars/WSTS/PDFs/1-0_Cisco_FBonomi_ConnectedVehicles.pdf
24. Bonomi F, Milito R, Zhu J, Addepalli S (2012) Fog computing and its role in the internet of things,
ACM Press, p 13. https://doi.org/10.1145/2342509.2342513. URL http://dl.acm.org/citation.cfm?
doid=2342509.2342513
25. Bonomi F, Milito R, Natarajan P, Zhu J (2014) Fog computing: a platform for Internet of Things
and analytics. In: Bessis N, Dobre C (eds) Big data and Internet of Things: a roadmap for smart
environments, vol 546, Springer International Publishing, Cham, pp 169–186. https://doi.org/10.
1007/978-3-319-05029-4_7. URL http://link.springer.com/10.1007/978-3-319-05029-4_7
26. Gazis V, Leonardi A, Mathioudakis K, Sasloglou K, Kikiras P, Sudhaakar R (2015) Components of
fog computing in an industrial Internet of Things context. In: IEEE, pp 1–6. https://doi.org/10.1109/
SECONW.2015.7328144. URL http://ieeexplore.ieee.org/document/7328144/
27. Abdullahi I, Arif S, Hassan S (2015) Ubiquitous shift with information centric network caching using
fog computing. In: Phon-Amnuaisuk S, Au TW (eds) Computational intelligence in information
systems, vol 331, Springer International Publishing, Cham, pp 327–335. https://doi.org/10.1007/
978-3-319-13153-5_32. URL http://link.springer.com/10.1007/978-3-319-13153-5_32
28. Skala K, Davidovic D, Afgan E, Sovic I, Sojat Z (2015) Scalable distributed computing hierarchy:
cloud, fog and dew computing. Open J Cloud Comput (OJCC) 2(1):9–00063
29. Mahmud R, Kotagiri R, Buyya R (2018) Fog computing: a taxonomy, survey and future directions.
In: Di Martino B, Li K-C, Yang LT, Esposito A (eds) Internet of everything, Springer, Singapore,
pp 103–130. https://doi.org/10.1007/978-981-10-5861-5_5. URL http://link.springer.com/10.1007/
978-981-10-5861-5_5
30. Saharan KP, Kumar A (2015) Fog in comparison to cloud: a survey. Int J Comput Appl 122(3):10–
12. https://doi.org/10.5120/21679-4773. URL http://research.ijcaonline.org/volume122/number3/
pxc3904773.pdf
31. Yi S, Li C, Li Q (2015) A survey of fog computing: concepts, applications and issues, ACM
Press, pp 37–42. https://doi.org/10.1145/2757384.2757397. URL http://dl.acm.org/citation.cfm?
doid=2757384.2757397
32. Yi S, Hao Z, Qin Z, Li Q (2015) Fog computing: platform and applications. In: IEEE, pp 73–78.
https://doi.org/10.1109/HotWeb.2015.22. URL http://ieeexplore.ieee.org/document/7372286/
33. Razouk W, Sgandurra D, Sakurai K (2017) A new security middleware architecture based on fog
computing and cloud to support IoT constrained devices, ACM Press, pp 1–8. https://doi.org/10.
1145/3109761.3158413. URL http://dl.acm.org/citation.cfm?doid=3109761.3158413
34. Alrawais A, Alhothaily A, Hu C, Cheng X (2017) Fog computing for the Internet of Things: security
and privacy issues. IEEE Internet Comput 21(2):34–42. https://doi.org/10.1109/MIC.2017.37. URL
http://ieeexplore.ieee.org/document/7867732/
35. Roman R, Lopez J, Mambo M (2018) Mobile edge computing, Fog et al.: a survey and analysis of
security threats and challenges. Future Gener Comput Syst 78:680–698. https://doi.org/10.1016/j.
future.2016.11.009. URL https://linkinghub.elsevier.com/retrieve/pii/S0167739X16305635

123
Fog computing: from architecture to edge computing and... 2099

36. Zhao P, Tian H, Fan S, Paulraj A (2018) Information prediction and dynamic programming-based
RAN slicing for mobile edge computing. IEEE Wirel Commun Lett 7(4):614–617. https://doi.org/
10.1109/LWC.2018.2802522. URL https://ieeexplore.ieee.org/document/8281474/
37. Krner M, Runge TM, Panda A, Ratnasamy S, Shenker S (2018) Open carrier interface: an open
source edge computing framework. In: Proceedings of the 2018 workshop on networking for emerging
applications and technologies—NEAT ’18, ACM Press, Budapest, Hungary, pp 27–32. https://doi.
org/10.1145/3229574.3229579. URL http://dl.acm.org/citation.cfm?doid=3229574.3229579
38. Syamkumar M, Barford P, Durairajan R (2018) Deployment characteristics of “The Edge” in
mobile edge computing. In: Proceedings of the 2018 workshop on mobile edge communications—
MECOMM’18, ACM Press, Budapest, Hungary, pp 43–49. https://doi.org/10.1145/3229556.
3229557. URL http://dl.acm.org/citation.cfm?doid=3229556.3229557
39. Yu W, Liang F, He X, Hatcher WG, Lu C, Lin J, Yang X (2018) A survey on the edge computing
for the Internet of Things. In: IEEE access, vol 6, pp 6900–6919. https://doi.org/10.1109/ACCESS.
2017.2778504. URL http://ieeexplore.ieee.org/document/8123913/
40. Jeong S, Simeone O, Kang J (2018) Mobile edge computing via a UAV-mounted cloudlet: optimization
of bit allocation and path planning. IEEE Trans Veh Technol 67(3):2049–2063. https://doi.org/10.
1109/TVT.2017.2706308. URL http://ieeexplore.ieee.org/document/7932157/
41. Rahmani AM, Gia TN, Negash B, Anzanpour A, Azimi I, Jiang M, Liljeberg P (2018) Exploiting
smart e-health gateways at the edge of healthcare Internet-of-Things: a fog computing approach.
Future Gener Comput Syst 78:641–658. https://doi.org/10.1016/j.future.2017.02.014. URL http://
linkinghub.elsevier.com/retrieve/pii/S0167739X17302121
42. Vora J, Tanwar S, Tyagi S, Kumar N, Rodrigues JJPC (2017) FAAL: fog computing-based patient
monitoring system for ambient assisted living. In: IEEE, pp 1–6. https://doi.org/10.1109/HealthCom.
2017.8210825. URL http://ieeexplore.ieee.org/document/8210825/
43. Fakeeh KA (2016) Privacy and security problems in fog computing. Commun Appl Electron 4:7
44. Taneja M, Davy A (2016) Resource aware placement of data analytics platform in fog computing. Pro-
cedia Comput Sci 97:153–156. https://doi.org/10.1016/j.procs.2016.08.295. URL http://linkinghub.
elsevier.com/retrieve/pii/S1877050916321111
45. Singh S, Chana I (2015) QoS-aware autonomic resource management in cloud computing: a system-
atic review. ACM Comput Surv 48(3):1–46. https://doi.org/10.1145/2843889. URL http://dl.acm.
org/citation.cfm?doid=2856149.2843889
46. Souza VB, Masip-Bruin X, Marin-Tordera E, Ramirez W , Sanchez S (2016) Towards distributed ser-
vice allocation in fog-to-cloud (F2c) scenarios. In: IEEE, pp 1–6. https://doi.org/10.1109/GLOCOM.
2016.7842341. URL http://ieeexplore.ieee.org/document/7842341/
47. Gupta M (2017) Fog computing pushing intelligence to the edge. Int J Sci Technol Eng 3(8):5
48. Zhao H, Li X (2013) Resource management in utility and cloud computing, SpringerBriefs in Com-
puter Science, Springer New York, New York. https://doi.org/10.1007/978-1-4614-8970-2. URL
http://link.springer.com/10.1007/978-1-4614-8970-2
49. Kameda H, Li J, Kim C, Zhang Y (1997) Optimal load balancing in distributed computer systems,
telecommunication networks and computer systems, Springer London. https://doi.org/10.1007/978-
1-4471-0969-3. URL http://link.springer.com/10.1007/978-1-4471-0969-3
50. Kopparapu C (2002) Load balancing servers, firewalls, and caches. Wiley, New York
51. Bittencourt LF, Rana OF (2017) Mobility-aware application scheduling in fog computing. IEEE
Cloud Comput 4:26–35
52. Etemad M, Aazam M, St-Hilaire M (2017) Using DEVS for modeling and simulating a fog comput-
ing environment. In: IEEE, pp 849–854. https://doi.org/10.1109/ICCNC.2017.7876242. URL http://
ieeexplore.ieee.org/document/7876242/
53. Aazam M, Zeadally S, Harras KA Offloading in fog computing for IoT: review, enabling technologies,
and research opportunities. Future Gener Comput Syst. https://doi.org/10.1016/j.future.2018.04.057.
URL http://linkinghub.elsevier.com/retrieve/pii/S0167739X18301973
54. Rayes A, Salam S (2017) Fog computing defining. In: Internet of Things from hype to reality, Springer
International Publishing, Cham, pp 139–164. https://doi.org/10.1007/978-3-319-44860-2_6. URL
http://link.springer.com/10.1007/978-3-319-44860-2_6
55. Klas GI (2015) Fog computing and mobile edge cloud gain momentum open fog consortium, ETSI
MEC and cloudlets

123
2100 S. P. Singh et al.

56. Deng R, Lu R, Lai C, Luan TH (2015) Towards power consumption-delay tradeoff by workload allo-
cation in cloud-fog computing. In: IEEE, pp 3909–3914. https://doi.org/10.1109/ICC.2015.7248934.
URL http://ieeexplore.ieee.org/document/7248934/
57. Dolui K, Datta SK (2017) Comparison of edge computing implementations: fog computing, cloudlet
and mobile edge computing. In: IEEE, pp 1–6. https://doi.org/10.1109/GIOTS.2017.8016213. URL
http://ieeexplore.ieee.org/document/8016213/
58. Hu P, Dhelim S, Ning H, Qiu T (2017) Survey on fog computing: architecture, key technologies,
applications and open issues. J Netw Comput Appl 98:27–42. https://doi.org/10.1016/j.jnca.2017.
09.002. URL http://linkinghub.elsevier.com/retrieve/pii/S1084804517302953
59. Fog computing in the internet of things (2017) Intelligence at the edge, 1st edn. Springer, New York
60. Toosi AN, Son J, Buyya R (2018) Clouds-pi: a low-cost raspberry-pi based testbed for software-
defined-networking in cloud data centers. ACM SIGCOMM Comput Commun Rev 7:1–11
61. Wang K, Shen M, Cho J, Banerjee A, Van der Merwe J, Webb K (2015) MobiScud: a fast mov-
ing personal cloud in the mobile network, ACM Press, pp 19–24. https://doi.org/10.1145/2785971.
2785979. URL http://dl.acm.org/citation.cfm?doid=2785971.2785979
62. Han B, Gopalakrishnan V, Ji L, Lee S (2015) Network function virtualization: challenges and oppor-
tunities for innovations. IEEE Commun Mag 53(2):90–97. https://doi.org/10.1109/MCOM.2015.
7045396. URL http://ieeexplore.ieee.org/document/7045396/
63. Vinueza Naranjo PG, Baccarelli E, Scarpiniti M Design and energy-efficient resource manage-
ment of virtualized networked Fog architectures for the real-time support of IoT applications. J
Supercomput. https://doi.org/10.1007/s11227-018-2274-0. URL http://link.springer.com/10.1007/
s11227-018-2274-0
64. Oueis J, Strinati EC, Barbarossa S (2015) The fog balancing: load distribution for small cell cloud com-
puting. In: IEEE, pp 1–6. https://doi.org/10.1109/VTCSpring.2015.7146129. URL http://ieeexplore.
ieee.org/document/7146129/
65. De Vleeschauwer D, Robinson DC (2011) Optimum caching strategies for a telco CDN. Bell Labs
Tech J 16(2):115–132. https://doi.org/10.1002/bltj.20506. URL http://ieeexplore.ieee.org/lpdocs/
epic03/wrapper.htm?arnumber=6770158
66. Pooranian Z, Shojafar M, Naranjo PGV, Chiaraviglio L, Conti M (2017) A novel distributed fog-based
networked architecture to preserve energy in fog data centers. In: IEEE, pp 604–609. https://doi.org/
10.1109/MASS.2017.33. URL http://ieeexplore.ieee.org/document/8108808/
67. Gupta P, Goyal MK, Gupta N (2015) Reliability aware load balancing algorithm for content delivery
network. In: Satapathy SC, Govardhan A, Raju KS, Mandal JK (eds) Emerging ICT for bridging the
future—proceedings of the 49th annual convention of the computer society of India (CSI), vol 337,
Springer International Publishing, Cham, pp 427–434. https://doi.org/10.1007/978-3-319-13728-
5_48. URL http://link.springer.com/10.1007/978-3-319-13728-5_48
68. Zhou J, Qiao Y (2015) Low-peak-to-average power ratio and low-complexity asymmetrically clipped
optical orthogonal frequency-division multiplexing uplink transmission scheme for long-reach passive
optical network. Opt Lett 40(17):4034. https://doi.org/10.1364/OL.40.004034. URL https://www.
osapublishing.org/abstract.cfm?URI=ol-40-17-4034
69. Nag A, Payne DB, Ruffini M (2016) N:1 protection design for minimizing olts in resilient dual-homed
long-reach passive optical network. J Opt Commun Netw 8(2):93. https://doi.org/10.1364/JOCN.8.
000093. URL https://www.osapublishing.org/abstract.cfm?URI=jocn-8-2-93
70. Dixit A, Lannoo B, Colle D, Pickavet M, Demeester P (2015) Delay models in ethernet long-reach pas-
sive optical networks. In: IEEE, pp 1239–1247. https://doi.org/10.1109/INFOCOM.2015.7218499.
URL http://ieeexplore.ieee.org/document/7218499/
71. De Andrade M, Buttaboni A, Tornatore M, Boffi P, Martelli P, Pattavina A (2015) Optimization of
long-reach TDM/WDM passive optical networks. Opt Switch Netw 16:36–45. https://doi.org/10.
1016/j.osn.2014.11.001. URL http://linkinghub.elsevier.com/retrieve/pii/S157342771400126X
72. Liu Y, Guo L, Yu C, Yu Y, Wang X (2014) Planning of survivable long-reach passive
optical network (LR-PON) against single shared-risk link group (SRLG) failure. Opt Switch
Netwo 11:167–176. https://doi.org/10.1016/j.osn.2013.06.001. URL http://linkinghub.elsevier.com/
retrieve/pii/S1573427713000404
73. Truong NB, Lee GM, Ghamri-Doudane Y (2015) Software defined networking-based vehicular adhoc
network with fog computing. In: IEEE, pp 1202–1207. https://doi.org/10.1109/INM.2015.7140467.
URL http://ieeexplore.ieee.org/document/7140467/

123
Fog computing: from architecture to edge computing and... 2101

74. He X, Ren Z, Shi C, Fang J (2016) Cloud/fog networking in the internet of vehicles. China Commun
13:140–149
75. Din S, Paul A, Ahmad A, Ahmed SH, Jeon G, Rawat DB (2018) Hierarchical architecture for 5g based
software-defined intelligent transportation system. In: IEEE INFOCOM 2018-IEEE Conference on
Computer Communications Workshops (INFOCOM WKSHPS), IEEE
76. Sheetal J Architecture of 5g technology in mobile communication. In: Proceedings of 18th IRF
International Conference, 11th January
77. Brown D, Mather D, Shaddock RN, Weeks WA, Franckx J, Erreygers JJJM (2018) Single line passive
optical network converter module. US Patent 9,900,108 (Feb. 20)
78. Chakraborty P (2018) Design of passive optical network for hospital management. Asian J Converg
Technol 4(I)
79. Mahmud R, Kotagiri R, Buyya R (2018) Fog computing: a taxonomy, survey and future directions.
In: Internet of everything, Springer, pp 103–130
80. Mercian A, McGarry MP, Reisslein M (2013) Offline and online multi-thread polling in long-reach
pons: a critical evaluation. J Lightwave Technol 31(12):2018–2028. https://doi.org/10.1109/JLT.2013.
2262766. URL http://ieeexplore.ieee.org/document/6515602/
81. Townsend PD, Talli G, MacHale EK, Antony C (2008) Long reach PONs, COIN 2008. In: 7th
International Conference on Optical Internet, pp 1–200000
82. Helmy A, Krishna N, Nayak A (2018) On the feasibility of service composition in a long-reach pon
backhaul. In: 2018 International Conference on Optical Network Design and Modeling (ONDM),
IEEE, pp 41–46
83. Helmy A, Nayak A (2018) Toward parallel edge computing in long-reach pons. J Opt Commun Netw
10(9):736–748
84. Arbelaez A, Mehta D, Sullivan OB, Quesad L (2018) Parallel constraint-based local search: an appli-
cation to designing resilient long-reach passive optical networks. In: Handbook of parallel constraint
reasoning, Springer, pp 633–665
85. Dastjerdi A, Gupta H, Calheiros R, Ghosh S, Buyya R (2016) Fog computing: principles, architectures,
and applications. In: Internet of Things, Elsevier, pp 61–75. https://doi.org/10.1016/B978-0-12-
805395-9.00004-6. URL http://linkinghub.elsevier.com/retrieve/pii/B9780128053959000046
86. Stojmenovic I, Wen S, Huang X, Luan H (2016) An overview of fog computing and its security issues:
an overview of fog computing and its security issues. Concurr Comput Pract Exp 28(10):2991–3005.
https://doi.org/10.1002/cpe.3485. URL http://doi.wiley.com/10.1002/cpe.3485
87. Dastjerdi AV, Gupta H, Calheiros RN, Ghosh SK, Buyya R (2016) Fog computing: principles, archi-
tectures, and applications. In: Internet of Things, Elsevier, pp 61–75
88. Chiang M, Zhang T (2016) Fog and iot: an overview of research opportunities. IEEE Internet Things
J 3(6):854–864
89. More P (2015) Review of implementing fog computing. Int J Res Eng Technol 4(06):335–338
90. Lin CC, Yang JW (2018) Cost-efficient deployment of fog computing systems at logistics centers in
industry 4.0. IEEE Trans Ind Inf. https://doi.org/10.1109/TII.2018.2827920
91. Jia G, Han G, Wang H, Wang F (2018) Cost aware cache replacement policy in shared last-level cache
for hybrid memory based fog computing. EnterpInf Syst 12(4):435–451
92. Sarkar S, Chatterjee S, Misra S (2018) Assessment of the suitability of fog computing in the context
of internet of things. IEEE Trans Cloud Comput 6(1):46–59
93. Song Z, Duan Y, Wan S, Sun X, Zou Q, Gao H, Zhu D (2018) Processing optimization of typed
resources with synchronized storage and computation adaptation in fog computing. Wireless Commun
Mob Comput. https://doi.org/10.1155/2018/3794175
94. He S, Cheng B, Wang H, Xiao X, Cao Y, Chen J (2018) Data security storage model for fog computing
in large-scale iot application. In: IEEE INFOCOM 2018-IEEE Conference on Computer Communi-
cations Workshops (INFOCOM WKSHPS), IEEE, pp 39–44
95. Bi Y, Han G, Lin C, Deng Q, Guo L, Li F (2018) Mobility support for fog computing: an sdn approach.
IEEE Commun Mag 56(5):53–59
96. Roig PJ, Alcaraz S, Gilly K, Juiz C (2018) Study on mobility and migration in a fog computing
environment. In: 22nd International Conference Electronics, IEEE, pp 1–6
97. Zhang P, Liu JK, Yu FR, Sookhak M, Au MH, Luo X (2018) A survey on access control in fog
computing. IEEE Commun Mag 56(2):144–149

123
2102 S. P. Singh et al.

98. Thota C, Sundarasekar R, Manogaran G, Varatharajan R, Priyan M (2018) Centralized fog computing
security platform for iot and cloud in healthcare system. In: Exploring the Convergence of Big Data
and the Internet of Things, IGI Global, pp 141–154
99. Wang B, Chang Z, Zhou Z, Ristaniemi T (2018) Reliable and privacy-preserving task recomposition
for crowdsensing in vehicular fog computing. In: IEEE 87th Vehicular Technology Conference (VTC
Spring), IEEE, pp 1–6
100. Guan Y, Shao J, Wei G, Xie M (2018) Data security and privacy in fog computing. IEEE Netw 99:1–6
101. Matt C (2018) Fog computing. Bus Inf. Syst Eng 60(4):351–355
102. Shi C, Ren Z, Yang K, Chen C, Zhang H, Xiao Y, Hou X (2018) Ultra-low latency cloud-fog computing
for industrial internet of things. In: 2018 IEEE Wireless Communications and Networking Conference
(WCNC), IEEE, pp 1–6
103. Mahmud R, Ramamohanarao K, Buyya R Latency-aware application module management for fog
computing environments. In: ACM Transactions on Internet Technology (TOIT)
104. Mahmud R, Srirama SN, Ramamohanarao K, Buyya R (2018) Quality of experience (QoE)-aware
placement of applications in fog computing environments. J Parallel Distrib Comput. https://doi.org/
10.1016/j.jpdc.2018.03.004
105. Chekired DA, Khoukhi L, Mouftah HT (2018) Industrial IoT data scheduling based on hierarchical
fog computing: a key for enabling smart factory. IEEE Trans Ind Inf 14(10):4590–4602. https://doi.
org/10.1109/TII.2018.2843802
106. Kiani A, Ansari N, Khreishah A Hierarchical capacity provisioning for fog computing. arXiv preprint
arXiv:1807.01093
107. Naqvi SAA, Javaid N, Butt H, Kamal MB, Hamza A, Kashif M (2018) Metaheuristic optimization
technique for load balancing in cloud-fog environment integrated with smart grid. In: International
Conference on Network-Based Information Systems, Springer, pp 700–711
108. Hussain MM, Alam MS, Beg MS (2019) Feasibility of fog computing in smart grid architectures.
In: Proceedings of 2nd International Conference on Communication, Computing and Networking,
Springer, pp 999–1010
109. Okay FY, Ozdemir S (2018) A secure data aggregation protocol for fog computing based smart
grids. In: 2018 IEEE 12th International Conference on Compatibility, Power Electronics and Power
Engineering (CPE-POWERENG), IEEE, pp 1–6
110. Lyu L, Nandakumar K, Rubinstein B, Jin J, Bedo J, Palaniswami M (2018) PPFA privacy preserving
fog-enabled aggregation in smart grid. IEEE Trans Ind Inf. https://doi.org/10.1109/TII.2018.2803782
111. Ling CW, Datta A, Xu J (2018) A case for distributed multilevel storage infrastructure for visual
surveillance in intelligent transportation networks. IEEE Internet Comput 22(1):42–51
112. Cao Y, Hou P, Brown D, Wang J, Chen S (2015) Distributed analytics and edge intelligence: pervasive
health monitoring at the era of fog computing. In: Proceedings of the 2015 Workshop on Mobile Big
Data, ACM, pp 43–48
113. Gia TN, Jiang M, Rahmani A-M, Westerlund T, Liljeberg P, Tenhunen H (2015) Fog computing in
healthcare internet of things: a case study on ecg feature extraction. In: 2015 IEEE International
Conference on Computer and Information Technology; Ubiquitous Computing and Communi-
cations; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing
(CIT/IUCC/DASC/PICOM), IEEE, pp 356–363
114. Aazam M, Huh E-N (2015) E-hamc: leveraging fog computing for emergency alert service. In: 2015
IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom
Workshops), IEEE, pp 518–523
115. Ballas C, Marsden M, Zhang D, O’Connor NE, Little S (2018) Performance of video processing at the
edge for crowd-monitoring applications. In: 2018 IEEE 4th World Forum Internet Things (WF-IoT).
https://doi.org/10.1109/WF-IoT.2018.8355170
116. Hong K, Lillethun D, Ramachandran U, Ottenwälder B, Koldehofe B (2013) Mobile fog: a program-
ming model for large-scale applications on the internet of things. In: Proceedings of the Second ACM
SIGCOMM Workshop on Mobile Cloud Computing, ACM, pp 15–20
117. Zhu X, Chan DS, Hu H, Prabhu MS, Ganesan E, Bonomi F (2015) Improving video performance
with edge servers in the fog computing architecture. Intel Technol J 19(1):202–224
118. Bonomi F, Milito R, Zhu J, Addepalli S (2012) Fog computing and its role in the internet of things.
In: Proceedings of the First Edition of the MCC Workshop on Mobile Cloud Computing, ACM, pp
13–16

123
Fog computing: from architecture to edge computing and... 2103

119. Grover J, Jain A, Singhal S, Yadav A (2018) Real-time vanet applications using fog computing.
In: Proceedings of First International Conference on Smart System, Innovations and Computing,
Springer, pp 683–691
120. Stojmenovic I, Wen S (2014) The fog computing paradigm: scenarios and security issues. In: 2014
Federated Conference on Computer Science and Information Systems (FedCSIS), IEEE, pp 1–8
121. Bonomi F, Milito R, Natarajan P, Zhu J (2014) Fog computing: a platform for internet of things and
analytics. In: Big Data and Internet of Things: A Roadmap for Smart Environments, Springer, pp
169–186
122. Liu J, Li J, Zhang L, Dai F, Zhang Y, Meng X, Shen J (2018) Secure intelligent traffic light control
using fog computing. Future Gener Comput Syst 78:817–824
123. Choo KKR, Lu R, Chen L, Yi X (2018) A foggy research future: advances and future opportunities
in fog computing research. Future Gener Comput Syst 78:677–679
124. Tran VL, Islam A, Kharel J, Shin SY (2018) On the application of social internet of things with fog
computing: a new paradigm for traffic information sharing system. In: 2018 IEEE 6th International
Conference on Future Internet of Things and Cloud (FiCloud), IEEE, pp 349–354
125. Rao YS, Sree KB (2018) A review on fog computing: conceptual live Vm migration framework,
issues, applications and its challenges. Int J Sci Res Comput Sci Eng Inf Technol 3(1)
126. Garg S, Singh A, Batra S, Kumar N, Yang LT (2018) Uav-empowered edge computing environment
for cyber-threat detection in smart vehicles. IEEE Netw 32(3):42–51
127. Li L, Ota K, Dong M (2018) Deep learning for smart industry: efficient manufacture inspection system
with fog computing. IEEE Trans Ind Inf 14(10)
128. Tortonesi M, Govoni M, Morelli A, Riberto G, Stefanelli C, Suri N (2018) Taming the IoT data
deluge: an innovative information-centric service model for fog computing applications. Future Gen
Comput Syst. https://doi.org/10.1016/j.future.2018.06.009
129. Raja K, Krithika L (2016) Smart street light system. Autom Auton Syst 8(4):97–99
130. Wang S, Dey S (2012) Cloud mobile gaming: modeling and measuring user experience in mobile
wireless networks. ACM SIGMOBILE Mob Comput Commun Rev 16(1):10–21
131. Zhao Z, Hwang K, Villeta J (2012) Game cloud design with virtualized cpu/gpu servers and initial
performance results. In: Proceedings of the 3rd Workshop on Scientific Cloud Computing, ACM, pp
23–30
132. Yang L, Cao J, Yuan Y, Li T, Han A, Chan A (2013) A framework for partitioning and execution of data
stream applications in mobile cloud computing. ACM SIGMETRICS Perform Eval Rev 40(4):23–32
133. Qian Z, He Y, Su C, Wu Z, Zhu H, Zhang T, Zhou L, Yu Y, Zhang Z (2013) Timestream: reliable
stream computation in the cloud. In: Proceedings of the 8th ACM European Conference on Computer
Systems, ACM, pp 1–14
134. Nath SB, Gupta H, Chakraborty S, Ghosh SK A survey of fog computing and communication: current
researches and future directions. arXiv preprint arXiv:1804.04365
135. Yi S, Li C, Li Q (2015) A survey of fog computing: concepts, applications and issues. In: Proceedings
of the 2015 Workshop on Mobile Big Data, ACM, pp 37–42
136. Yi S, Hao Z, Qin Z, Li Q (2015) Fog computing: platform and applications. In: Third IEEE Workshop
on Hot Topics in Web Systems and Technologies (HotWeb), IEEE 2015, pp 73–78
137. Rahmani A-M, Thanigaivelan NK, Gia TN, Granados J, Negash B, Liljeberg P, Tenhunen H, Smart,
(2015) e-health gateway: bringing intelligence to internet-of-things based ubiquitous healthcare. sys-
tems. In: 2015 12th Annual Consumer Communications and Networking Conference (CCNC), IEEE,
pp 826–834
138. Mahmoud MM, Rodrigues JJ, Ahmed SH, Shah SC, Al-Muhtadi JF, Korotaev VV, De Albuquerque
VHC (2018) Enabling technologies on cloud of things for smart healthcare. IEEE Access 6:31950–
31967
139. Din S, Paul A, Guizani N, Ahmed SH, Khan M, Rathore MM (2017) Features selection model for
internet of e-health things using big data. In: GLOBECOM 2017—2017 IEEE Global Communica-
tions Conference, IEEE, pp 1–7
140. Varghese B, Wang N, Barbhuiya S, Kilpatrick P, Nikolopoulos DS (2016) Challenges and oppor-
tunities in edge computing. In: IEEE, pp 20–26. https://doi.org/10.1109/SmartCloud.2016.18. URL
http://ieeexplore.ieee.org/document/7796149/
141. Shenoy K, Bhokare P, Pai U (2013) FOG computing future of cloud computing. Int J Sci Res (IJSR)
4(6):55–56

123
2104 S. P. Singh et al.

142. Hao Z, Novak E, Yi S, Li Q (2017) Challenges and software architecture for fog computing. IEEE
Internet Comput 21(2):44–53. https://doi.org/10.1109/MIC.2017.26. URL http://ieeexplore.ieee.org/
document/7867731/
143. Varghese B, Wang N, Nikolopoulos DS, Buyya R (2017) Feasibility of fog computing. arXiv preprint
arXiv:1701.05451
144. Puthal D, Obaidat MS, Nanda P, Prasad M, Mohanty SP, Zomaya AY (2018) Secure and sustainable
load balancing of edge data centers in fog computing. IEEE Commun Mag 56(5):60–65
145. Wan J, Chen B, Wang S, Xia M, Li D, Liu C (2018) Fog computing for energy-aware load balancing
and scheduling in smart factory. IEEE Trans Ind Inf. https://doi.org/10.1109/TII.2018.2818932
146. Iorga M, Feldman L, Barton R, Martin MJ, Goren NS, Mahmoudi C (2018) Fog computing conceptual
model. Technical report
147. Aazam M, Zeadally S, Harras KA (2018) Deploying fog computing in industrial internet of things
and industry 4.0. IEEE Trans Ind. Inf 14(10):4674–4682
148. Comma-di L, Abdullaziz OI, Antevski K, Chundrigar SB, Gdowski R, Kuo P-H, Mourad A, Yen L-H,
Zabala A (2018) Opportunities and challenges of joint edge and fog orchestration. In: 2018 IEEE
Wireless Communications and Networking Conference Workshops (WCNCW), IEEE, pp 344–349
149. Chaudhary D, Bhushan K, Gupta B (2018) Survey on ddos attacks and defense mechanisms in cloud
and fog computing. Int J E-serv Mobile Appl (IJESMA) 10(3):61–83
150. Jiang Y, Huang Z, Tsang DH (2018) Challenges and solutions in fog computing orchestration. IEEE
Netw 32(3):122–129
151. Santos J, Vanhove T, Sebrechts M, Dupont T, Kerckhove W, Braem B, Van Seghbroeck G, Wauters
T, Leroux P, Latre S et al (2018) City of things: enabling resource provisioning in smart cities. IEEE
Commun Mag 56(7):177–183
152. Wu H-Y, Lee C-R, Energy efficient scheduling for heterogeneous fog computing architectures. In:
2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC), IEEE, pp
555–560
153. Mehta A, Elmroth E (2018) Distributed cost-optimized placement for latency-critical applications
in heterogeneous environments. In: 2018 IEEE International Conference on Autonomic Computing,
Trento, Italy, September 3–7, 2018, pp 121–130
154. Byers CC, Clarke JM, Salgueiro G (2018) Configuring heterogeneous computing environments using
machine learning. US Patent App. 15/390,921 (Jun. 28)
155. Cappiello C, Plebani P, Vitali M (2018) A data utility model for data-intensive applications in fog
computing environments. In: Fog computing, Springer, pp 183–202
156. Khan MA, Umer T, Khan SU, Yu S, Rachedi A (2018) Ieee access special section editorial: green
cloud and fog computing: energy efficiency and sustainability aware infrastructures, protocols, and
applications. IEEE Access 6:12280–12283
157. Qiao G, Leng S, Zhang K, He Y (2018) Collaborative task offloading in vehicular edge multi-access
networks. IEEE Commun Mag 56(8):48–54
158. Aazam M, Zeadally S, Harras KA (2018) Offloading in fog computing for IoT: review, enabling
technologies, and research opportunities. Future Gen Comput Syst 87:278–289. https://doi.org/10.
1016/j.future.2018.04.057
159. Zhang G, Shen F, Yang Y, Qian H, Yao W (2018) Fair task offloading among fog nodes in fog
computing networks. In: 2018 IEEE International Conference on Communications (ICC), IEEE, pp
1–6
160. Jošilo S, Dán G Decentralized fog computing resource management for offloading of periodic tasks.
In: Poster Presented at IEEE INFOCOM
161. Agarwal S, Yadav S, Yadav AK (2015) An architecture for elastic resource allocation in Fog. Com-
puting 6(2):7
162. Alrawais A, Alhothaily A, Hu C, Cheng X (2017) Fog computing for the internet of things: security
and privacy issues. IEEE Internet Comput 21(2):34–42
163. Tsugawa M, Matsunaga A, Fortes JA (2014) Cloud computing security: What changes with software-
defined networking? Secure cloud computing. Springer, Berlin, pp 77–93
164. Hu P, Ning H, Qiu T, Song H, Wang Y, Yao X (2017) Security and privacy preservation scheme of
face identification and resolution framework using fog computing in internet of things. IEEE Internet
Things J 4(5):1143–1155
165. Wang C, Wang Q, Ren K, Lou W (2010) Privacy-preserving public auditing for data storage security
in cloud computing. In: 2010 Proceedings, Infocom, IEEE, pp 1–9

123
Fog computing: from architecture to edge computing and... 2105

166. Basudan S, Lin X, Sankaranarayanan K (2017) A privacy-preserving vehicular crowdsensing-based

road surface condition monitoring system using fog computing. IEEE Internet Things J 4(3):772–782
167. Koo D, Hur J (2018) Privacy-preserving deduplication of encrypted data with dynamic ownership
management in fog computing. Future Gener Comput Syst 78:739–752
168. Ma L, Teymorian AY, Cheng X (2008) A hybrid rogue access point protection framework for com-
modity wi-fi. networks. In: The 27th Conference on Computer Communications INFOCOM 2008,
IEEE, pp 1220–1228
169. Modi C, Patel D, Borisaniya B, Patel H, Patel A, Rajarajan M (2013) A survey of intrusion detection
techniques in cloud. J Netw Comput Appl 36(1):42–57
170. Valenzuela J, Wang J, Bissinger N (2013) Real-time intrusion detection in power system operations.
IEEE Trans Power Syst 28(2):1052–1062
171. Qin Z, Li Q, Chuah M-C (2013) Defending against unidentifiable attacks in electric power grids.
IEEE Trans Parallel Distrib Syst 24(10):1961–1971
172. Cao N, Wang C, Li M, Ren K, Lou W (2014) Privacy-preserving multi-keyword ranked search over
encrypted cloud data. IEEE Trans Parallel Distrib Syst 25(1):222–233
173. Rial A, Danezis G (2011) Privacy-preserving smart metering. In: Proceedings of the 10th Annual
ACM Workshop on Privacy in the Electronic Society, ACM, pp 49–60
174. Wang L, Liu G, Sun L (2017) A secure and privacy-preserving navigation scheme using spatial
crowdsourcing in fog-based vanets. Sensors 17(4):668
175. Qin Z, Yi S, Li Q, Zamkov D (2014) Preserving secondary users’ privacy in cognitive radio networks.
In: 2014 Proceedings of INFOCOM, IEEE, pp 772–780
176. Wei W, Xu F, Li Q (2012) Mobishare: flexible privacy-preserving location sharing in mobile online
social networks. In: 012 Proceedings of INFOCOM„ IEEE, pp 2616–2620

123