Resource Allocation in Cloud Using the Adaptive Chronological

Based Exponential Lion Optimization algorithm

Cloud computing is a service-dependent framework that renders the on-demand access to the
resources for various applications. The main intention of the cloud computing is to allocate
the shared resources effectively and to maintain the profit of the cloud provider. In this paper,
an adaptive optimization method is introduced to allocate the resources optimally for cloud
computing. The adaptive chronological-based exponential lion is the modified algorithm,
which is developed through the integration of the adaptive concept in the exponential lion
optimization algorithm that inherits the advantages of exponential concept in Lion
optimization algorithm (LOA). The cloud provider allocates the resources effectively as
requested by the users using proposed adaptive exponential lion optimization (adaptive E
Lion) algorithm, which is based on the fitness function that is designed based on the
performance evaluation metrics, such as CPU utilization rate, memory utilization rate, and
frequency scaling factor. The proposed method is analysed under problem instances and
evaluated based on performance evaluation metrics. The proposed adaptive chronological
based exponential lion outclassed other comparative models, which acquired the profit, CPU
Utilization rate, and Memory Utilization rate values of 21.8705, 0.1032 and
0.00139,respectively.

Keywords: cloud computing, adaptive exponential lion optimization, resources, cloud

provider, network

11625
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636

1. Introduction

The emergence of cloud computing has an impact on Information Technology (IT)

industries, as they have helped developers and companies to overcome the lack of hardware
capacity by allowing the users to access on-demand resources through the Internet [1].
Information Technology (IT) industries compete against one another in providing more cost
efficient, reliable and powerful cloud services and they are striving to gain more benefits by
reshaping their business services. The service providers ensure their profits by providing good
QoS and maximum user satisfaction. The important role in cloud computing is the resource
allocation [5]. Due to its heterogeneous nature, the resource allocation in cloud computing is
based on two methods, namely system centric and user centric. The system centric is the
traditional method, which optimizes overall throughput of the system. The user-centric
method is based on QoS requirements thus, providing maximum utilization for the users. The
primary problem in cloud computing is to tackle the dynamic behaviour, such as the
availability of resources and workloads by maintaining good QoS constraints together with
maintaining an acceptable level of system performance [3].

The failure in resource allocation leads to loss of service and this can be prevented by
allocating the resources to the separate modules than the entire cloud framework [4]. The
resource allocation with both private and external cloud differ from one another, as the
resources utilized in the private cloud provides more profit than the resources utilized from
the external cloud [8]. These days, resource allocation using optimization based schemes have
gained importance. In this paper, the chronological E-Lion algorithm is modified using the
adaptive concept for effective resource allocation in the cloud. The constant parameters used
for evaluating the fertility of female lion are modified to attain precise results in the proposed
method. Million Instructions per Second (MIPS), frequency scaling factor profit, cost, CPU
utilization rate, and memory utilization rate are the cloud parameters used for evaluating the
fitness function.

The organization of the paper is as follows: section 1 describes the background of cloud
computing, Section 2 reviews the existing methods of cloud computing. Section 3 deliberates
the proposed method and section 4 describes the results and discussion of the proposed
method and finally, Section 5 concludes the paper.

2. Literature Survey

The review of the existing methods is presented in this section: Weiwei Xia et al. [2]
designed a joint resource allocation using evolutionary algorithms in heterogeneous mobile
cloud computing network. Although this method had maximum quality of service (QoS), such
as end-to-end response latency and the total utility of users, it had high operating cost. Siqian
Gong et al. [5] modelled an Adaptive Multivariable Control for Multiple Resource Allocation
of Service-based Systems in Cloud Computing. In this method, the resource allocation reacted
to resource demands and dynamic request workloads. This method had increased resource
utilization and operating costs, but the transition process was unstable, as it cannot adapt to
frequent changes in the controller. Shahin Vakilinia et al. [1] developed a VM placement to
minimize total power consumption of cloud DCs. In this method, the complexity and
computation time were reduced by cut and solve based algorithm and call back method, but it
required high processing power consumption. Gongzhuang Peng et al. [6] modelled a
11626
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
mathematical model to represent the complex resource scheduling problem in a multi-tenant
computing environment and an improved multi-objective genetic algorithm was developed
based on elitist archive and k-means approaches. This method would not analyze node price
and priority weight.

2.1 Challenges
The challenges of the research are enlisted below:
∙ The efficiency of the cloud model is decreased as most of the resource allocation
 schemes concentrate only on increasing the profit of the service provider. Hence, the
efficiency of the cloud model should be taken in to consideration [9].
∙ Most of the algorithms allocate only a single resource to the incoming task, but some
tasks require more than one resource [9].
∙ The resource allocation suffers as the demand of services increases and it is essential to
consider the workload of resources and the servers during optimization of the tasks
[10].
∙ Due to its diversified characteristics, the task assignment to the ICT resources, like
processing units, and storage devices faces difficulties due to the impact caused by
the dynamic availability and flexibility of resources [11].

3. Proposed adaptive chronological-E-Lion for resource allocation in the

cloud

Millions of users’store, retrieve and process their data with the help of cloud computing, as
the cloud handles numerous bytes of data per second. Thus, the allocation of these resources
to the user is more challenging. The cloud resources, such as physical machine, and virtual
machine are allocated based on the task. The entire cloud operation is controlled by the
resource allocation module and a small degradation in performance makes the cloud
environment inefficient. The crucial factor in the allocation of resources is to satisfy the
multiple task requirement of the consumer within the deadline and making more reasonable
allocation of the resources even if the cloud resources are exploited. The cloud service
provider, such as crucial factor uses its own resources whereas the Extra cloud (EC) utilizes
the private cloud during scarcity.

Figure (1) shows the cloud framework in the distributed platform. Initially, the computing
resources are requested by the consumer from the cloud. The incoming tasks are collected by
the private cloud through the user interface module. If the incoming tasks are higher than the
processing speed, the tasks are stored in the task buffer. The validity of the tasks is monitored
by the remote manager in the private cloud module. The accepted tasks from the remote
manager are processed through the scheduler. Through the resource monitor module, the
scheduler tracks the resources allocated for the tasks, details about the incoming tasks and the
available resources. For monitoring the processes done in the resource pool, the resource
monitor is directly connected with the resource pool. When all the resources are exploited in
the private cloud, the scheduler utilizes the resource from EC. The resources are collected
from EC based on pricing model, which is collected through the cloud interface in exchange
of price.

11627
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
 Applications

Private cloud environment Incoming tasks

External cloud 1

User interface

Resource pool
External cloud 2
Task buffer

Adaptive
Cloud consumers Request manager
Chronologic al based
E Lion Scheduler
algorithm
Cloud interface
Resource monitor External cloud E

Figure1. Cloud Framework for resource allocation in cloud

4. Resource allocation based on the proposed adaptive Chronological E

Lion algorithm

The procedure of the resource allocation scheme and the proposed adaptive Chronological
E-Lion algorithm is depicted in this section. The Adaptive Chronological E-Lion algorithm
optimization algorithm is developed to allocate the resources to the task. The resources are
distributed in the vast surface in distributed cloud computing and hence, the allocation of
resources is important and the tasks considered here are independent belonging to different
applications. The VMs are controlled by PM, which has its own CPU memory, speed and
frequency scaling parameters.

11628
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
 International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
In the distributive cloud framework, assume the set of cloud providers as, Cand the
numbers of cloud providers are assumed asC={C1,C2,...,Cn }; where nis the number of cloud
providers, whereinC1is the private cloud and { , ,..., } C2 C3 Cnare the external clouds. Let M
={M1,M2,...,Mm}is the set of virtual machine types for the execution of the corresponding
application tasks. Let Hbe the set of application requested by the user and it is given as, H
={h1,h2,...,hx}. The task of the user request application is set as, { } My my my mk S
= 1 2.
, ,...,

4.1 Task Flow

The availability and the deadline meeting criteria of the virtual machine Mare required for
the task allocation execution of the user application. Let us assume the task be s (y =1,2,3)
y,

deadline be J y∈1,2,3 ={1,2,3}and the run time of the task be Py∈1,2,3 ={1,1,1}. Initially, if there
are three tasks from the user, the task 1swith less cost is provided with resource from the
cloud. The resources are allocated by checking the availability of virtual machine in private
cloud and by checking the deadline of the task. If Mcan meet the deadline then the resource s
.
are allocated for the task 1

4.2 Fitness formulation

The factors, such as profit, cost, CPU utilization rate, memory utilization rate, MIPS, and
frequency scaling factor are used for evaluating the fitness computation. The fitness function
is derived as two functions for achieving better optimal resource allocation and it is
represented as,
12 Fitness function Efit =λ×E +(1−λ)×E(1)
1 2
where, λis the normalizing constant between 0 and 1, E and E are the two objective function
1
in fitness computation, in which E is allied with income and cost, and frequency scaling
2
factor and E is allied with CPU utilization rate, memory utilization rate, and MIPS. The
1 1
value of E should be maximum for best fitness function. The objective function E is
computed as,
E =(Income−Cost) 1(2) The income depends on frequency scaling factor, runtime and
price, is represented as,
m
p
=
∑ ∑ ** (3)
Income Gi D P j ij
i i
= jM
∈ ∈ 1 =1 pm

js
where, Gi, Djand Pijrepresents the frequency scaling factor, runtime, and price for the
resource allocation. The other factors, such as cost depends on frequency scaling factor,
runtime, and cost, and it is represented as,

11629
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
p m

=
∑ ∑ ** (4)
Cost Gi D Q j ij
i i
= jM
∈ ∈ 1 =1 pm

js

where, Qijis the cost allocation of the task.Similarly, the objective function E2is formulated
using CPU utilization rate, memory utilization rate, and MIPS, respectively. The function is
calculated by adding normalization function (1− λ)in the fitness and the function is
represented as,
()
∑ ∑ ()
⎥
⎥ ⎦⎤
=
∈ ∈=

⎢⎡
⎢⎣××
CPU mem MIPS
QQQ
p m
ij
× × ij ij

2(5)
E =
11 i RRR
js pm

i
CPU mem MIPS j M

where, CPU Qij, mem Qij, and MIPS Qijrepresents the number of CPUs, memory size, and MIPS
utilized during the allocation of the task. The termsRCPU, Rmem, and RMIPSrepresents the total
number of CPUs, memory size, and MIPS in a cloud provider.

4.3 Optimization driven resource allocation using the proposed adaptive

Chronological E-Lion algorithm

The main idea of the Lion algorithm is the survival of the fittest and it is inspired by the
social behaviour of the lion. The continuous optimization problems are solved by lion
optimization algorithm. The proposed Adaptive Exponential Lion algorithm increases the
profit of the cloud provider and the convergence problem is reduced by integrating the
adaptive nature, chronological theory, and the lion optimization algorithm. This section
explains the adaptive E- Lion algorithm for the optimal resource allocation in cloud
computing. The solution coding and fitness evaluation are also explained in this section.

a) Solution Encoding:At first, the resource is allocated for the task, which increases the profit
of the cloud provider by reducing the cost price. Figure 2 shows the solution encoding. In
solution encoding, the virtual machines are assigned with task, which has low cost price and
the task with high cost price are assigned at last.
s s s s s s s
1 2 3 4 5 6 7

s s s s s s s
2 1 3 6 4 7 5

Figure 2. Solution Encoding

11630
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
4.3.1 Algorithmic description of the proposed adaptive Chronological E-Lion algorithm

This section describes the optimization of the resource allocation for the proposed adaptive
E-Lion optimization algorithm. The Adaptive Chronological E-Lion algorithm is the
integration of adaptive and chronological E-Lion algorithm, whereas chronological E-Lion
algorithm is the integration of chronological concept and E-Lion algorithm, which is the
integration of EWMA [13] and LOA [12]. LOA increases the speed in the optimal solution
reachability and the convergence problem is defined clearly in the lion solution reach. The
algorithmic steps of the proposed adaptive E-Lion optimization algorithm are given below.

a) Pride generation: Let M V , F V, and N Vbe the pride generation of the male, female and
nomad lions and they are the arbitrary integers generated within the maximum and minimum
F
, V (i)
M N
, V (i) ; where i=1,2,3,...,I. The
limits. The arbitrary integers are given as, V (i)

fitness of M V , F Vand N Vis calculated using the fitness function, which helps in
understanding the feasibility of the lion pride. For the optimal solution detection, the
reference fitness is the fitness value of the male lion.

b) Evaluation of Fertility of female lion:After the pride generation, the fertility rate of the
male and the female lion are evaluated. The fertility of the male lion is evaluated by
initializing the laggardness rate and the fertility of the female lion is evaluated by sterility
rate. The updates based on the chronological and EWMA helps in evaluating the fertility of
the female lion. The proposed adaptive chronological E-Lion optimization algorithm modifies
the chronological E-Lion optimization algorithm with the adaptive optimization parameters.
The update equation of the proposed equation is expressed as,
1
 {2 ( 1) (1 ) ( ) (1 ) ( 1) (1 ) ( 1)}
F
Vb
γγγγγ ( )2
−−−
VbUVbUUVaUVa
F F
=+−−*−−**−−−*−
kk kk
v
−γ
2 is the fertility of the
U female lion based
F

(6)
where, Uis the constant value in EWMA, V (b) v
−
Vb
F
γ
refers to the updated fertility rate of the E-Lion at
on E-Lion at iteration b , and ( +1)
k

iteration(b +1). Thus, the actual fertility evaluation equation of female lion is given
as, γ γ α γ
( 1) ( ) (0.1 0.05)( ( ) ( )) Vk b Vk b l2 V b l1Vk b
+ = + − − (7) where, 1land 2lare the constant for evaluating the fertility of female lion.
In the proposed adaptive chronology E-lion algorithm, the constant parameters gets modified
as,
⎜⎛ ⎟
⎜⎝− ⎟ ⎠⎞
F γ
−
Vb
lF ()
k
1
exp V b (8)
=−−
γ
k
( 1)
l = −l(9) (1 ) 2 1

c) Crossover and Mutation for gender clustering: The new solutions are generated based
on crossover and mutation operation after the evaluation of fertility of the male and the
female lions. A maximum of four cubs are yielded during the crossover of the male and the
female lion irrespective of the gender and the crossover operation is performed by a crossover
mask. The crossover operation is represented as,
ηαγ
V (a) =WaV +WaV(10) where, Wais the crossover mask, with length aand η Vis the
resultant cub after the crossover operation. The new cub lions formed after the crossover
operation by mutation
11631
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
operation, is represented as,new Vη _and the clustered male cubs and female cubs are
represented as η _α Vand η _γ V, respectively.

d) Formulating the cub growth function: As the growth depends on the age, the cub growth
is denoted as mutation function. The fitness of the cub lion is determined and the lion with the
highest fitness value is determined as the new male and female. The growth function is
formulated as the factor for the territorial takeover.

e) Territorial defense against nomad lions:In the search space, LA considers two nomad
lions, which has a chance to take over the territory. For the evaluation, the nomad lions are
represented as,V (i)
and V (i)
N1
N2
. The male lion defends its territory against the nomad lions
and the territory can be taken only if the male lion has less fitness than the nomad lion.
Therefore, the fitness of both the male and the nomad lion has consequences in the defense
mechanism. At first, the nomad lion V (i)
N1
has the survival fight with the male lion and the
lion wins with better fitness. After that, nomad lion V (i)
N2
is eligible to fight with the lion.
The eligible nomad lion takes over the pride group is given as,
E ⎟ ( ( ) ( ))
⎟ ⎠⎞
σ 2
Fitness V Fitness
⎜ d
⎜ ⎝⎛ 2 max ,
V
σσ
=(11) exp σ

() 12

max , ( )1
dd Fitness V

where, d1is the Euclidean distance measure between N1 Vand α V ,2 dis the Euclidean distance
measure between N1 Vand α V .
f) Update based on territorial takeover: The fitness of the cub is considered and after some
iteration, if the fitness of the cub exceeds the fitness of the lion, the cub takes over the
territory.
g) Termination:Finally, after the maximum iteration, the solution with the best fitness is
considered as the optimal solution. In the cloud environment, the final solution is the
optimally allocated with the resources.

4. Results and discussion

The result and discussion of the proposed adaptive E-Lion optimization approach for the
resource allocation in the cloud computing are discussed in this section.

4.1 Experimental setup

The proposed adaptive E-Lion approach for the optimized resource allocation in the cloud
computing environment is performed on a PC with Intel core i3 processor, and 2 GB memory
using the windows 7 operating system. The simulation of the proposed method is
implemented in JAVA.
4.2 Performance metrics
The performance metrics considered for the evaluation of the proposed adaptive E-Lion
approach over the existing approach are profit, CPU Utilization rate, and Memory Utilization
rate.

11632
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
4.3 Comparative methods
The performance of the proposed Adaptive Chronological based ELION is compared with
the existing methods, such as Particle swarm optimization(PSO) [14], Self-Adaptive Learning
PSO (SLPSO) [15], LION [13], Exponential-Lion (ELION), and Chronological-based ELION
that is the integration of the chronological concept in LION.

4.4 Comparative analysis

In this section, performance of the proposed optimization approach over the problem
instance-1 consisting of five applications is evaluated. The VM instance type is randomly
selected from virtual machines, as requested by each application from private or external
clouds. The deadline of each application is restricted between the random integer of 1 h and 7
h in order to restrict the search space.

a) Profit Evaluation:Figure 3 shows the profit evaluation curve for the problem instance.
The problem instance is a small instance, which attains optimal solution within a reasonable
time. The profit attained by the proposed Adaptive Chronological-based E-Lion and the
existing methods, such as PSO, SLPSO, LION, ELION, Chronological based E-Lion at 50th
iteration are 21.8705, 5.9376, 7.2617, 18.2992, 19.2536, and 19.2949, respectively.

Figure 3. Profit Evaluation curve for Problem Instance 1

b) CPU Utilization rate evaluation: Figure 4 shows the CPU utilization rate evaluation
curve for the problem instance-1. The CPU utilization rate for the proposed Adaptive
Chronological-based E-Lion and the existing methods, such as PSO, SLPSO, LION, ELION,
Chronological-based ELION, at 50th iteration are 0.1032, 0.0020, 0.0025, 0.0031, 0.0032, and
0.1030, respectively. The CPU utilization rate for the proposed Adaptive Chronological-based
ELION is maximal.

11633
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 - 11636

Figure 4. CPU Utilization rate Evaluation curve for Problem Instance 1

c) Memory Utilization rate Evaluation: Figure 5 shows Memory utilization rate evaluation
curve for the problem instance-1. The Memory utilization rate for the proposed Adaptive
Chronological-based ELION and the existing methods, such as PSO, SLPSO, LION, ELION,
Chronological-based ELION, at 50th iteration are 0.00139, 5.56E-04, 0.00103, 0.00104,
0.0013, and 0.0014, respectively. In all the iterations, the profit of the cloud provider
increases as the size of the memory increases in private cloud.
 Figure 5. Memory utilization rate evaluation curve for problem instance
1

11634
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -
11636
5. Conclusion

In this paper, an optimization algorithm was introduced integrating the adaptive concept in
the exponential lion optimization algorithm for the optimal allocation of resources. The
resources are allocated effectively by the cloud provider using proposed adaptive exponential
lion optimization algorithm. The fitness function of the adaptive E-Lion is designed using the
factors, such as profit, CPU Utilization rate, and Memory Utilization rate. The proposed
method is analyzed under problem instances. The performance of the proposed system was
compared with the existing approaches, such as PSO, SLPSO, LION, ELION, and
Chronological-based ELION using the performance measures. The proposed method
outperformed the existing method with maximum profit of 21.8705, CPU Utilization rate of
0.1032, and Memory Utilization rate of 0.00139, respectively.

References
[1] Vakilinia, S., Heidarpour, B., & Cheriet, M. (2016). Energy Efficient Resource Allocation in Cloud Computing
Environments. IEEE Access, 4, 8544–8557.
[2] Xia, W., & Shen, L. (2018). Joint resource allocation using evolutionary algorithms in heterogeneous mobile
cloud computing networks. China Communications, 15(8), 189–204.
[3] Samimi, P., Teimouri, Y., & Mukhtar, M. (2016). A combinatorial double auction resource allocation model in
cloud computing. Information Sciences, 357, 201–216.
[4] S. H. H. Madni, M. S. A. Latiff, Y. Coulibaly, and S. M. Abdulhamid, “Recent advancementsin resource
allocation techniques for cloud computing environment: a systematic review,” Cluster Comput., vol. 20, no.
 3, pp. 2489–2533, 2017.
[5] Gong, S., Yin, B., Zheng, Z., & Cai, K. (2019). Adaptive Multivariable Control for Multiple Resource
Allocation of Service-based Systems in Cloud Computing. IEEE Access, 1–1.
[6] Peng, G., Wang, H., Dong, J., & Zhang, H. (2018). Knowledge-Based Resource Allocation for Collaborative
Simulation Development in a Multi-Tenant Cloud Computing Environment. IEEE Transactions on Services
Computing, 11(2), 306–317.
[8] Z. Li, T. Chu, I. V. Kolmanovsky, X. Yin, and X. Yin, “Cloud resource allocation for cloud-based automotive
applications,” Mechatronics, no. December 2016, pp. 1–10, 2017.
[9] W. Wei, X. Fan, H. Song, X. Fan, and J. Yang, “Imperfect Information Dynamic Stackelberg Game Based
Resource Allocation Using Hidden Markov for Cloud Computing,” IEEE Trans. Serv. Comput., vol. 1374,
no. c, pp. 1–1, 2016.
[10] S. Vakilinia, “Energy efficient temporal load aware resource allocation in cloud computing datacenters,” J.
Cloud Comput., vol. 7, no. 1, pp. 1–24, 2018.
[11] Z. Cao, J. Lin, C. Wan, Y. Song, Y. Zhang, and X. Wang, “Optimal Cloud Computing Resource Allocation for
Demand Side Management,” IEEE Trans. Smart Grid, pp. 1–13, 2016.
[12] Michael S. Saccucci , Raid W. Amin & James M. Lucas, "Exponentially weighted moving average control
schemes with variable sampling intervals", Communications in Statistics - Simulation and Computation, vol.
21, no.3, pp. 627-657, 1992.
[13] B. Rajakumar, "The Lion‫׳‬s Algorithm: a new nature-inspired search algorithm", Procedia Technology, vol .6,
pp. 12–135, 2012.
[14] SurajPandey, Linlin Wu, SiddeswaraMayura Guru, RajkumarBuyya, "A Particle Swarm Optimization-based
Heuristic for Scheduling Workflow Applications in Cloud Computing Environments", The Cloud Computing
and Distributed Systems (CLOUDS) Laboratory, University of Melbourne.
[15] XingquanZuo, Guoxiang Zhang, and Wei Tan, "Self-Adaptive Learning PSO-Based Deadline Constrained
Task Scheduling for Hybrid IaaS Cloud", IEEE Transactions on Automation Science And Engineering, vol.
11, no. 2, 2014.

11635
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC
International Journal of Advanced Science and Technology
Vol. 29, No. 3, (2020), pp. 11625 -

Authors
11636

Mr. J. Devagnanam is currently working as Assistant Professor in

Thiruthangal Nadar College, Chennai. He is also the Research
Scholar of Bharathiar University, Coimbatore, India. He has more
5
than 1 years of experience in academic and industry
. He attended
and presented many papers in National and International
conferences and published his papers in springer and I
nderscience
journals and also in UGC approved journals.

Dr.N.M.Elango is currently working as Associate Professor in

School of Information Technology and Engineering, Vellore
 Institute of Technology, Vellore, Tamil Nadu 632014, India. He is
being Research Guide in various Institutions and Universities. He

35
has more than 30 years of experience in academic and published
more than National and International Journals.His research

is well known academicianas we l a

interests include Computer Networks, Cloud Computing, and Image
Processing. He l sresearcher.

11636
ISSN: 2005-4238 IJAST
Copyright ⓒ 2020 SERSC