2021 3rd International Conference on Signal Processing and Communication (ICPSC) | 13 – 14 May 2021 | Coimbatore

The World of Communication & Computing Platform

In Research Perspective: Opportunities and
Challenges
2021 3rd International Conference on Signal Processing and Communication (ICPSC) | 978-1-6654-2864-4/20/$31.00 ©2021 IEEE | DOI: 10.1109/ICSPC51351.2021.9451711

V Balaji Kasyap N. Jayapandian

School of Engineering and Technology School of Engineering and Technology
CHRIST (Deemed to be University), India CHRIST (Deemed to be University), India
[email protected] [email protected]

Abstract—Computing paradigms are introduced for solving This traditional type of computing was called as “Serial
complex problems by analyzing, designing and implementing by computing”. So, the concept of “Parallel Computing” was
complex systems. Computing can be defined as the effective use introduced [2]. Instead of a single computational device
of computer or computer technology to solve tasks that are goal multiple of them working parallel together reduced the time
oriented. Computing is used in development of producing [3]. Advancement in parallel computing made “Distributed
scientific studies, building intelligent systems, channeling Computing” possible for better results [4]. Emerging
different media for communication. Over the last few years, technologies require the integration of the most recent trends
internet became so popular which lead to the increase in and scientific applications for high performance and maximum
computer processing capacity, data storage and communication
utilization [5]. This cannot be achieved using a super computer
with one another. Computing has evolved from one technology to
at a place or a cluster of them. Hence being inspired by the
another in its field and formed a robust framework over the
years. In this paper a survey on different computing paradigms electrical grids a replica of computers was designed by the
like evergreen computing is cloud computing, to deal with basic computer scientists in the mid 1990’s [6]. This replica is called
scheduling is grid computing, for multi task handing is parallel as computational power grid whose name evolved to “grid
computing, to handle smart phone data’s that is mobile computing” [7]. However Cloud Computing is the predecessor
computing, cluster computing, and distributed computing is of Grid Computing in which data storage devices and software
carried out. These technologies improved the way computing computing services were added in addition to computational
functions and made it easier to the computer world. The power [8]. Cloud computing being the recent most
applications and research issues of the most of the computing development in computing is flourishing and hence need to be
paradigms are discussed in this article. The recent research issues signified [9]. According to a survey, 80% of all the
in computing platform are scheduling and security. The organizations are likely to use the cloud based services by
scheduling is dealing with data processing from one computing 2025.
platform to other computing device. Security is one of the
important research issues. II. CLOUD COMPUTING
Almost any organization dealing with digital data requires a
Keywords— Cloud Computing; Cluster Computing; Distributed
Computing; Grid Computing; Mobile Computing; Parallel
place like servers, processing units for the sake of storing and
Computing; processing. But setting up a new hardware or software in a big
amount is a money induced process [10]. And this is where
I. INTRODUCTION Cloud Computing takes over, by enabling the clients to store
and make use of the services one that has to offer on a pay and
The evolution of computing to parallel, distributed, grid and
use basis. You can run an operating system, use software, play
cloud solved the problems one by one that the traditional
games, stream video, business process management and many
computing faced. This can lead to the expansion of different
more without actually running on local devices [11]. All you
technologies like Internet of things, Artificial Intelligence,
need is a good and uninterrupted internet connection and can
Intelligent Applications (I-Apps), Dev-opps, and many more.
be accessed from any part of the world. The cloud started to be
In the olden days, software’s were written in such a way that
a buzz by 2010 and became at its peaks by 2020. The main
the instructions execute one after another [1]. The reason
reason is why cloud computing emerged fast is due to the low
behind can be, computing power being likely less as compared
cost and high quality services given by the cloud hosts [12].
to the present due to various factors including the hardware.
This breakthrough has happened with the virtualization
The functioning of the software was based not on “Divide and
powering the cloud. Here the term Cloud is just another
rule” policy but “Divide and execute” policy. Problems to be
metaphor referring the internet or centralized servers. But,
solved using computers increased as they involved in the lives
where is this cloud located? Public Cloud is famous
of common man. That’s when developers had a convolution as
organizations like public cloud service providers are the
the time consuming was more with the traditional computing.

978-1-6654-2864-4/21/$31.00 ©2021 IEEE 289

Authorized licensed use limited to: California State University Fresno. Downloaded on July 01,2021 at 18:41:18 UTC from IEEE Xplore. Restrictions apply.
 2021 3rd International Conference on Signal Processing and Communication (ICPSC) | 13 – 14 May 2021 | Coimbatore

providers of a public internet service by not asking to pay for III. GRID COMPUTING
hardware or software that are being used. User can use the Grid Computing is the descendent and an evolution from
services offered by the cloud hosts but lack the authorization to Distributed Computing. It was inspired by the electrical power
manage, change or control the cloud. While using the services, grid. The user needs not to know the geographical location of
user feels as it is running on the local computer (it actually runs the power plant while using the electricity [15]. All one cares is
on remote cloud centers).Thanks for the virtualization for if they are getting the required power. The same applies to the
making this possible. Private cloud is when an organization grid computing users as well. In short way grid computing
sets up a cloud for its own use or a third party vendor hosting it consists of a number of heterogeneous or homogeneous hosts,
to different organizations differently to make use then it is each with several computational resources. Grid computing has
called as a private cloud [13]. It is not that contrast with the evolved from the preexisting technologies like distributed
public cloud. While the public cloud vendor’s offers service to computing, the Internet, web services, most of the
many, private cloud is just for a particular organization. The cryptographic techniques alongside powering up with the
resources are not shared with other organizations but for the virtualization technology [16]. The grid computing works
sole purpose of own. But being not cost effective is a major around these four building blocks. They are user, resource
setback. Hybrid cloud is cloud bursting or Hybrid cloud is broker, grid information service and resources. The user
making the use of both public and private clouds together [14]. requests for executing a task is given to the grid broker. Then
An organization uses its own private cloud for computing and the broker splits the job as required by the user into several
also the public cloud if the computing requirements do not tasks and submits them to various resources that are available.
match up with private cloud. This ensures the better results in The grid information service takes a note of information
terms of computing. Companies who use hybrid cloud are, regarding the resources that took part in scheduling. The
Netflix, Ola, Uber, Airbnb and ellipsis. Moreover, after resources are what user wants and the user is someone who
availability and location the main feature of the cloud is to uses resources. In technical terms the major components that
provide services. The classification of cloud based on the made grid computing are schedulers, load balancing and grid
services they offer are, IaaS (Infrastructure as a Service): portals [17]. Schedulers are basically the task managers who
Computing resources like servers, databases, data centers, collect the necessary resources for a particular task execution,
networking are offered by the Cloud providers to the users on distributing the tasks to work on an operation parallels,
pay per use basis. reserving the resources, keeping track of the task execution and
its status. The load balancing is to eliminate delays in
processing and deadlocks, the load balancing units are used.
These load balancers are built upon the schedulers and resource
managers. The grid portals can be identified as the access
points to an internet system. It basically provides a structure or
a domain for the user to utilize the services that the grid has to
offer. Also it plays a major role in keeping track of network
applications and association of other users. Grid Portals
manages the entry and authentication of the users using the
information from job scheduling, file transfer and information
resources. Grid computing is a five layered structure in which
its protocols and structures are present. The fabric layer is
placed at the bottom of the grid architecture. The resources that
need to be shared are present in this layer; like processors,
Figure 1: Cloud Computing Different Service catalogues, network resources, and sensors. Connectivity layer
The figure 1 is discussing the different cloud computing is the protocols required for authentication and communication
services. As these resources are dynamic in nature, they are is present in this layer. Resource layer is relevant duties related
feasible to use and are resized, managed, divided in a way how to network infrastructure like negotiation, accounting, payment,
the user wants. Amazon S3 is an important service, IBM’s is monitoring, and controlling are taken care in this layer. The
provides blue cloud. The Rackspace cloud is some companies protocol called Grid Resource Access and Management
that offer this type of services. PaaS (Platform as a Service): (GRAM) is used to allocate the computational resources [18].
Along with the attributes of IaaS some software servers and The collective layer helps in keeping the multiple resources in
applications are added additionally. Users can build, test, and sync and its management. Application Layer is the top most
deploy the web applications from scratch to fully functional layer is the application layer where all the user friendly
without investing additionally on any hardware, software or applications and API’s are built based on the above protocols.
infrastructural cost. Windows Azure, Google App engine and The basic origin of grid computing is dealing with different
many other offers these types of services. SaaS (Software as a kind of computing that is high throughput computing is used to
Service): Here the software is the main priority and is being deal with higher server data. The modern world big data is also
given as a service to the users. Any user based applications, handling larger amount of data that basic mechanism is can
software programs can be run on a cloud which is well from data intensive computing. The grid computing, utilization
maintained and managed by the cloud provider. Google apps of resources are taken are in an efficient manner, work
(Google Docs, Sheets, Classroom) make a better permeability is reduced; huge amount of data is accessible both
understanding. locally and globally.

290

Authorized licensed use limited to: California State University Fresno. Downloaded on July 01,2021 at 18:41:18 UTC from IEEE Xplore. Restrictions apply.
 2021 3rd International Conference on Signal Processing and Communication (ICPSC) | 13 – 14 May 2021 | Coimbatore

IV. DISTRIBUTED COMPUTING workstations which could be cost effective. Cluster Computing
Distributed computing or Distributed systems are a network means a cluster of processors associated to a shared network
of individual computer systems and is equipped with a and it is a single supercomputer to compute [24]. Cluster
middleware called as distributed machine software [19]. The computing is used for high availability and high performance
main working mechanism of this distributed computing is to which increases computational power than a single common
handle the system resources and manage the client and server computer. The main objective of cluster computing is to make
resources. The introduction of distributed computing increased the effective use of parallel processing of multiple computers
the connectivity of computers, fault management, performance, connected in a single local area network. As the nodes increase
and resource sharing [20]. In distributed computing, any in the network the work load on each node reduces and
problem is divided into multiple parts and each part is processing speed increases. The components of cluster
processed by a computational device or unit. The execution of computing include multiple nodes (High Performance
the operation is handled by the distributed applications present Computers), operating systems, high performance networks,
on all the computers in the network. The client and server fast and reliable communication protocols and services, trusted
model is base structure of distributed computing. Clients are middleware. Moreover, clusters are divided into three. High
basically the computer (End Nodes) that are connected in the obtainability clusters is can also be called as failover clusters
network and use the services provided by the service providers. as, if a single node in the network fails to work, still the
Server is kind of a remote computer that shares the data to the services offered by the cluster can be presented. But there
clients based on the request to the client. Basically a client might be a performance reduction because of the node that is
requests for the data and the server provides the data to the unavailable. The main feature of these clusters is to provide
client unless it is present in the database. Along with the client- continuous availability of the services that the cluster has to
server architecture, there are a few more architectures included offer. Computing services like network file distribution,
into the Distributed computing model. They are peer to peer, database management, and business tasks can be achieved by
multilayered, and service oriented architecture. High availability clusters. Load balancing clusters is
distributing the incoming requests among multiple nodes that
V. PARALLEL COMPUTING are present in the cluster and that have the same programs
running. Load balancing clusters combined with high
Parallel computing involves much of a hardware availability clusters can help increase the availability,
environment and the way it has been built. The early scalability of data resources like web, mail, ftp services. Hence
technology VLSI (Very Large Scale Integration) has been in service quality is enhanced. In general high-performance
the background of parallel computing. Here one or more clusters are used to solve complex problems in a very short
computers are interconnected in such a way that they start note time are known as reliable performance groups. This
working together on a specific task [21]. Earlier computer cluster system is used in research labs, media and
software’s were written for serial execution which involved a entertainment, financial services, artificial intelligence and
lot of time to solve a problem. Only one instruction is allowed machine learning. Also these clusters are used in treating
to execute at a time making the real life problem solving to be diseases like cancer, diabetes for the accurate results of the
nearly impossible. In parallel computing a problem is broken patient in a short time. The major benefit of the cluster based
down into different small problems and is again divided into computing is to link the one node to other node very quick at
simpler instructions. Though it is not feasible to develop the same time it is available anywhere and anytime. It also
parallel software nowadays it was a major contribution to the achieved in cluster computing along with cost efficiency.
technological field in the mid 1980’s. Fifth generation
computers owns this technology. The two computer field’s VII. MOBILE COMPUTING
computational science and video game technology combined
made parallel computing possible. A hardware is introduced The mobile computing is a technology that is used to
which plays a major role in parallel computing called transfer data in the form of text, audio or video through a node
GPU(Graphical Processing Unit).It is a power processor without actually using a physical network. Which means the
involving parallel processing by increasing performance than data is transferred among wireless enabled devices. The device
that of a normal CPU [22]. A point to be noted is that it is not need not be at a fixed location to share the data which makes it
restricted to just graphical implementations but many more. to be called computing on the move [25]. In mobile computing
the data can be accessed from anywhere at any time. The
VI. CLUSTER COMPUTING mobile computing architecture is based on three elements. First
one is Mobile Communication it is used to communicate one
Before getting into cluster computing, let us consider the mobile system to other. The second one is mobile hardware, it
idea of supercomputers and workstations [23]. A workstation is is dealing with mother board and microprocessor. Final
a multi-core processor machine which helps out a single user component is mobile software; it is heart of mobile
with intensive tasks, which a normal desktop or a computer communication. Mobile communication is the backbone for
could not do. And a supercomputer is a multiple user machine working of the mobile computing. It makes sure of establishing
designed specifically for highest intensity tasks like weather a reliable and consistent communication. The communicating
forecasting, seismic data, oil exploration, and quantum device can be a computer or a mobile among which data is
mechanics. Setting up a supercomputer for computation is a transferred. Mobile communication can be of these three forms.
very costly and risky job. Hence the concept of cluster The mobile and wired fixed and wired and mobile and
computing was introduced. With proper programming and wireless. The laptops are best examples. The mobile devices
maintenance supercomputers are established using a cluster of

291

Authorized licensed use limited to: California State University Fresno. Downloaded on July 01,2021 at 18:41:18 UTC from IEEE Xplore. Restrictions apply.
 2021 3rd International Conference on Signal Processing and Communication (ICPSC) | 13 – 14 May 2021 | Coimbatore

are not mobile and are fixed at a position with physical two types of verification are involved. That is semantic
network for communication. In mobile and wireless structure verification and syntactic structure verification. Both
configuration, without the use of any wired device, nodes can semantic structure verification and syntactic structure
communicate with each other. Mobile hardware is the verification are to be considered for the effective grid
communicating nodes that share and access the service of workflow. Dead activity is the lack of synchronization and
mobility. The software component is used to execute the some of the main syntactic structure inconsistencies.
mobile device. It plays a major role in working of the mobile Performance parameters defined are to be in match with the
hardware by acting as an interface just like an operating users requirement in a grid workflow. Some user defined
system. parameters include average resource allocation, average
activity completion time, average capacity utilization rate,
VIII. RESEARCH ISSUES IN COMPUTING DOMAIN average activity queuing time. Hence a comparison between
Computing has made a promising future to the system performance parameters and user defined parameters is
technological field over the past few decades due to the to be carried out. Many simulation tools are used to achieve
progression of internet. But there has been so much of research this. Many methods like Markovian chain, queuing theory are
that is still being carried out by the scientists in different fields also used. However the Morkovian chain technique may be
of computing. Though there has been solutions to most of the mostly time consuming process. Resource verification is done
problems there are a ton more to be addressed and the issues in order to check whether there is any conflict between
arises every day. Cloud Computing has been a used activities in terms of resources. Sometimes two or more
extensively in IT sector by many companies and business activities may require the same machine at same time interval.
enterprises. But it cannot be considered as fully developed So there is a requirement of conducting a statistical analysis
technology as there are many issues to be addressed even for wither the activities actually need the machine or not.
today due to its extensive usage. The issues that are majorly Table 1: Computing Usage in Different Application
considered are security issues, data issues, performance issues,
and hardware design issues. Virtual Machine Migration is
considered by most of the cloud providers as a solution. Applications
Sometimes, there can be a maintenance issue or a technical
glitch that requires a server to be shut down. In such cases the Different
Edu
entire VM with it its operating system and data applications are Computin
cati Gover Financ Enterta Bankin
transferred from one machine to another. This method is called g
on nment e inment g
as Virtual Machine Migration. The advantages of VM Sect Sector Sector Sector Sector
migration are workload migration, elasticity and scalability, or
resilience. Maximizing resource utility and minimizing the
Cloud
energy consumption is a challenging task in an environment
like cloud computing. Hence the functioning of a VM is done Computin √ √ √ √ √
on unused or underused servers. This method is called as g
Server Consolidation. The problem of consolidating the server Grid
optimally is called as NP-hard optimization problem to which Computin √ √ × × √
many heuristics approaches have been introduced. Information g
Security is also one of the main issues that need to be Distribute
addressed in Cloud Computing. Hence the server provides d
√ √ √ √ √
should maintain Confidentiality and audit ability to the users. Computin
The confidentiality means secure accessing and transferring the g
data from one computing device to other. To achieve Parallel
confidentiality many cryptographic protocols have been Computin √ × × √ ×
developed. To check whether the security settings of the g
applications are proper and not tampered, audit ability is Mobile
considered. Remote attestation techniques are used to achieve Computin × √ √ √ √
audit ability. Non forgeable security system analysis is done by
g
the trusted platform modules. But for Cloud Computing where
migration of VMs is very normal extensively, remote Cluster
attestation may not be the best solution. Hence, trust Computin × × √ √ ×
mechanisms including many encryption and decryption g
techniques are built in the architecture of the cloud. Deadlock
and livelock, temporal violation, resource conflict is to be The table 1 gives a brief study on where different
mitigated for the proper workflow in grid computing. Some of computing paradigms are used exclusively in different areas.
the research issues like structure verification, performance Cloud computing is almost used in most of the sectors in
verification, authorization, cost verification, and temporal different ways. Cloud Computing in education sector is making
verification are to be considered while dealing with the both student’s and teacher’s life easier by enabling the online
workflow in grid computing. Usually structure verification is classes even in the Covid-19 pandemic. By this students can
the process of verifying the consistency of the structure. Here even submit their assignments, and do almost many things just

292

Authorized licensed use limited to: California State University Fresno. Downloaded on July 01,2021 at 18:41:18 UTC from IEEE Xplore. Restrictions apply.
 2021 3rd International Conference on Signal Processing and Communication (ICPSC) | 13 – 14 May 2021 | Coimbatore

like a physical classroom. Government is most likely to protect [9] Tabrizchi, H., & Rafsanjani, M. K. A survey on security challenges in
data and strives to reduce the cost. It make easier for the cloud computing: issues, threats, and solutions. The journal of
supercomputing, 76(12), 9493-9532. (2020)
government to reduce relying upon IT companies providing
[10] Slimani, S., Hamrouni, T., & Charrada, F. B. Service-oriented
infrastructure. The resources are not limited and so does the replication strategies for improving quality-of-service in cloud
updates of the applications are much easier. For the sake of computing: a survey. Cluster Computing, 24(1), 361-392. (2021)
batch processing, data storage and data analytics many banks [11] Sun, X., Gao, B., Zhang, Y., An, W., Cao, H., Guo, C., & Sun, W.
use cloud computing which is much more flexible and reliable. Towards delivering analytical solutions in cloud: Business models and
Content acquisition, editing, broadcasting, playing video games technical challenges. In 2011 IEEE 8th International Conference on e-
have been drastically improved under the entertainment Business Engineering (pp. 347-351). IEEE. (2011)
industry due to the cloud computing technology. Globus toolkit [12] Jayapandian, N., & Md Zubair Rahman, A. M. J. Secure deduplication
for cloud storage using interactive message-locked encryption with
that has been a part of grid computing is used for sharing convergent encryption, to reduce storage space. Brazilian Archives of
resources, accessing data, providing security, and e-learning. Biology and Technology, 61. (2018)
[13] Basmadjian, R., De Meer, H., Lent, R., & Giuliani, G.Cloud computing
IX. CONCLUSION and its interest in saving energy: the use case of a private cloud. Journal
This article is discussing different computing paradigms of Cloud Computing: Advances, Systems and Applications, 1(1), 1-25.
(2012)
like cloud computing, grid computing, distributed computing,
parallel computing, cluster computing and, mobile computing [14] Clemente-Castelló, F. J., Nicolae, B., Mayo, R., & Fernández, J. C.
Performance model of mapreduce iterative applications for hybrid cloud
are discussed with brief overview. With the information of the bursting. IEEE Transactions on Parallel and Distributed Systems, 29(8),
survey a brief study is provided. Making two or more 1794-1807. (2018)
computers to complete a given task is the parallel computing. [15] Jain, M., Kain, P., Gupta, D., & Rodrigues, J. J. The Role of Intelligent
Distributed computing is similar to parallel computing where Grid Technology in Cloud Computing. Applications of Cloud
different instances of a problem run simultaneously. By Computing: Approaches and Practices, 83. (2020)
connecting two or more computers and make it work as a [16] N. Jayapandian, A. M. J. M. Z. Rahman, M. Koushikaa and S.
single computer for handling intense workloads is the cluster Radhikadevi, "A novel approach to enhance multi level security system
using encryption with fingerprint in cloud," 2016 World Conference on
computing. Virtualization is the basis of cloud computing and Futuristic Trends in Research and Innovation for Social Welfare (Startup
grid computing. Use of the grids in grid computing was Conclave), Coimbatore, India, 2016, pp. 1-5, doi:
inspired by the electrical grids. Cloud computing has been 10.1109/STARTUP.2016.7583903. (2016)
introduced in recent years which offers a wide range of web [17] Paznikov, A. A. Optimizing Logical Structures of Decentralized
services with a cheaper cost and is yet integrated with Schedulers in Multicluster Systems and Computational Grids. In 2020
virtualization. International Multi-Conference on Industrial Engineering and Modern
Technologies (FarEastCon) (pp. 1-6). IEEE. (2020)
REFERENCES [18] Sinha, P., Aeishel, G., & Jayapandian, N. Computational Model for
Hybrid Job Scheduling in Grid Computing. In Intelligent
Communication Technologies and Virtual Mobile Networks (pp. 387-
[1] Li, L., Lin, Y. L., Zheng, N. N., Wang, F. Y., Liu, Y., Cao, D., & 394). Springer, Cham. (2019)
Huang, W. L. Artificial intelligence test: a case study of intelligent [19] El-Sayed, H., Sankar, S., Prasad, M., Puthal, D., Gupta, A., Mohanty,
vehicles. Artificial Intelligence Review, 50(3), 441-465. (2018) M., & Lin, C. T. Edge of things: The big picture on the integration of
[2] Rozenberg, G., Bäck, T., & Kok, J. N. (Eds.). Handbook of natural edge, IoT and the cloud in a distributed computing environment. IEEE
computing (pp. 461-477). Springer Berlin Heidelberg. (2021) Access, 6, 1706-1717. (2017)
[3] Kuang, L., Gong, T., OuYang, S., Gao, H., & Deng, S. Offloading [20] Bandopadhaya, S., Dey, R., & Suhag, A. Integrated healthcare
decision methods for multiple users with structured tasks in edge monitoring solutions for soldier using the internet of things with
computing for smart cities. Future Generation Computer Systems, 105, distributed computing. Sustainable Computing: Informatics and
717-729. (2020) Systems, 26, 100378. (2020)
[4] Xu, Y., Liu, H., & Long, Z. A distributed computing framework for [21] Cicirelli, F., Giordano, A., & Mastroianni, C. Analysis of Global and
wind speed big data forecasting on Apache Spark. Sustainable Energy Local Synchronization in Parallel Computing. IEEE Transactions on
Technologies and Assessments, 37, 100582. (2020) Parallel and Distributed Systems, 32(5), 988-1000. (2020)
[5] Tychalas, D., & Karatza, H. A scheduling algorithm for a fog computing [22] Healey, R., Dowers, S., Gittings, B., & Mineter, M. J. (Eds.) Parallel
system with bag-of-tasks jobs: Simulation and performance evaluation. processing algorithms for GIS. CRC Press. (2020)
Simulation Modelling Practice and Theory, 98, 101982. (2020) [23] Geng, X., Zhang, H., Zhao, Z., & Ma, H. Interference-aware
[6] Fernández, Á., Fernández, C., Miguel-Dávila, J. Á., & Conde, M. Á. parallelization for deep learning workload in GPU cluster. Cluster
Integrating supercomputing clusters into education: a case study in Computing, 23(4), 2689-2702. (2020)
biotechnology. The Journal of Supercomputing, 1-24. (2020) [24] Abbasi, M., Shokrollahi, A., Khosravi, M. R., & Menon, V. G. High-
[7] Sinha, P., Aeishel, G., & Jayapandian, N. Computational Model for performance flow classification using hybrid clusters in software defined
Hybrid Job Scheduling in Grid Computing. In Intelligent mobile edge computing. Computer Communications, 160, 643-660.
Communication Technologies and Virtual Mobile Networks (pp. 387- (2020)
394). Springer, Cham. (2019) [25] Liao, B., Ali, Y., Nazir, S., He, L., & Khan, H. U. (2020). Security
[8] Abraham, J. J., Sunny, C., Assisi, A., & Jayapandian, N. Cloud analysis of IoT devices by using mobile computing: a systematic
Virtualization with Data Security: Challenges and Opportunities. In literature review. IEEE Access, 8, 120331-120350. (2020)
International conference on Computer Networks, Big data and IoT (pp.
865-872). Springer, Cham. (2018)

293

Authorized licensed use limited to: California State University Fresno. Downloaded on July 01,2021 at 18:41:18 UTC from IEEE Xplore. Restrictions apply.