Cloud Computing Task Scheduling Algorithm Based on Modified Genetic Algorithm

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 08 | Aug 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 403
Cloud Computing Task Scheduling Algorithm Based on Modified
Genetic Algorithm
Ms. Manorma Kori 1, Mr. Rajneesh Pachouri2, Mr. Anurag Jain3
M.Tech Research Scholar Department of Computer Science Engineering AIST, Sagar
Assistant Professor, Department of Computer Science Engineering AIST, Sagar
Assistant Professor, Department of Computer Science Engineering AIST, Sagar
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The design of cloud computing allows for
scalable computing. The cloud components can be used to
process many requests and handle them in a timely manner.
The cloud aids in the handling of multiple requests and the
secure management of user data. There are components that
locate a suitable architecture and so offer communication
bonding between components. The key communication
components available in the cloud are virtual machines, data
centres, and user bases. As a result, it is always necessary to
handle many requests, assign the appropriatevirtualmachine
to the input request, and then provide the fastest response
time possible. There are numerous methods for balancing the
load on a virtual machine. The data locality preservation is
rigorous in the original article, which makes load balancing
across nodes a difficult task when the approach is used. a
heuristic method. Most range-queriable cloud storage
currently uses a combination of neighbour item exchangeand
neighbour migration methods, which has a highoverheadand
sluggish convergence. Algorithms like Round robin, throttle,
and other VM allocation aid with machine allocation, butonly
to a limited extent. While progress is being made toward
better virtual machine allocation, finding the best possible
allocation is constantly needed in order to increase
performance. In this paper, a Rule-based threshold heuristic
technique is described as an algorithm. This is the algorithm
that combines the many characteristicsofvirtualmachines, as
well as their statuses, to determine the optimum virtual
machine for request allocation. The method issimulated using
the Cloud Analyst simulation tool, andacomparison isdone by
applying an existing algorithm to several topologies.
Key Words: Cloud Load Balancing, Data sharing,
Virtualization, Heuristic Approach, Bully search, Localization,
Dynamic Allocation.
1. INTRODUCTION
“Cloud Environment is a platform that combines usable
entities from many sources and makes them available as a
service on the WWW Internet platform on an as-needed
basis, freeing users of the load of administering and
exchanging a dedicated complicated computing
infrastructure” [1]. “The availability of abundantly
provisioned given data management centres, as well as the
creation of elastic cloud infrastructures, open up new
applications and business models, and it may disrupt cloud
environment providers' operations” [2]. “Cloud computing
and data storage is a virtual platform that allowsforefficient
services through the internet. These on-demand IT useable
entities are efficiently generated and disposed of, auto-
completing utilising the various variables available
programmatic data UI, and invoicing is based on their
working and quantifiable component utilisation. Usable
entities are allotted in a traditional hosted environment
based on peak load needs” [3]. As the Cloud environment
data store and its computation emerge as a good way to
leverage available remote usable entitiess in a flexible, cost-
effective manner with its scaling way thanks to a usage-
based available here a cost model, security is one of the
available critical concerns that directly impact the adoption
working rate of the scenario Cloud paradigm [4]. System
virtualization, for example, hasbecomeextensivelyaccepted
to provide compute usable entities as a service, allowing the
dynamic spawning of virtual machines and their linked
nodes and communication infrastructureindata centres.[1].
When given by a provider to a group of set consumers, one
cloud service type known as software as a service (SaaS)has
attracted the attention of attackers who aim to exploit its
working weaknesses [5]. For security reasons, a defined VM
in the cloud is used with a proxy that replicates inbound
traffic to the devices and transmits it to the emulation
platform [6]. Below figure 1.1 shows discussion point from
cloud.
Figure 1 Complex Structure of Cloud
environment data store & its Computation.

1.1 LOAD BALANCING
Load balancing is a procedure that distributes workload
evenly among all available nodes in the cloud. By spreading
workloads between multiple nodes, this enhances overall
system performance. When resources are not utilised
effectively, they might get overheated, resulting in carbon
emissions. Carbon emissions can be reduced by utilising
resources wisely [4]. A few measures that can be used to
evaluate load balancing systems include performance,
scalability, response time, resource consumption, and fault
tolerance. These criteria enable us to determine whether the
approach or algorithm for balancing the given load is
adequate for balancing the load [5].
2. PROBLEM DEFINITION
According to the writing review, many methods are used to
do calculations, and different results are seen, such as
PRISM, SVM, and other different techniques for planning
procedures on a large number of arranged information
packages available dataset.
Following confirmation of the specific circumstance and the
available approach, a variety of shorts accompany the
existing booking computation, which serves as the
foundation for our investigation work. The checked focuses
are next, which were identified as an issue and were further
broken down and completed with improvements.
1. Previous strategies, such as PRISM and other booking
calculations for handling model age, have an evidentissue in
creating better results and information allotment strategies
when the virtual matchine network is broken. This method
produces a better result than the present system, but it also
requires improvement, which the proposed system
provides..
2. Previous approach Due to a lack of rules, nave based
characterization does not provide a superior enrollment
grouping, and hence a superior probability model cannot be
produced using the method.
3.In the past, dispersion was usedsincetheknowledgeofthe
points varied, determining the disadvantage of unexpected
substances in contrast to recommended tasks such as
heuristic hunt and circulation calculation.
As a result, in order to propose a better forecast model
utilising arrangement and further join approaches, it is
necessary to promote a scheme that contributes to
improving results and framework; in thiscase,ourproposed
procedure heuristic is to use conspire instead of the
traditional booking approach.
3. PROPOSED METHODOLOGY
As a result of our observations of the prior technique and its
drawbacks in various contexts and scenarios. Our research
presents a novel approach that is both productive and high-
value, resulting in improved computational results when
applied to a large number of datasets.Ourresearchproposes
a novel method, Heuristic Based Prediction Model that
employs a new logistic normal distribution technique to
establish a relationship between the topics as well as a
flexible environment for the entire process, resulting in a
superior data transmission prediction model. The proposed
algorithm is described below:
1. Loading of all available data and packets from
the created message that are involved in the
communication.
2. Loading the dataset's whole node dictionary
pair.
3. Execute the specific algorithm that the user has
chosen for continued execution, such as an
existing or developed algorithm.
4. If any single match is obtained, perform node
down and matching operations, and conclude
whether or not utilising the model for data
shifting is working.
5. If the system performs at least two or more
dictionary matches, do the model and match
action.
6. Keeping parameter-by-parameter data for the
historical model.
7. Observing the values and thus it effect accuracy
and efficiency for the complete scenario.
8. Exit .
Figure 2 General Structure of Load balancing in
Cloud Environment.

In this fig 3 tool and here as we load the dataset and verifies
the eligibility and taking their features for consideration or
not is the time taking process to identify and to load the
images and selection of password comesundertrainingtime
of a dataset, extracting the properties and making them in
process format is training time.
4. RESULT ANALYSIS
In this section, we'll go over the results that were achieved
after using the proposed technique instead of the old
method.
Parameters Description
Throughput: In general, throughput refers to the pace at
which something is produced or processed.
Computation Time: The amount of time it takes to finish a
process. Time to compute. The length of time necessary to
complete a computational process is known as computation
time (also known as running time). The computation time is
proportional to the number of rule applications when the
computation is represented as a sequence of rule
applications.
In the table presentbelowisa statistical comparison
of the values which are retrived as time taken by the
differentprocess algorithm,throughputandotherparameter
can be observe.
Table -1: Data distribution for different data packet.
The above table shows the amount of data values extracted
from the data and the algorithm used to extract them
Figure 4 Comparison of Line graph for technique
analysis
In the above graph, the x axis representsthedata fromwhich
posts were extracted for query processing for the specified
dataset, and the line graph is printed using the chart library
provided by Microsoft, allowing for easy further analysis.As
a result, the Heuristic based approach outperforms the best.
Our proposed approach outperforms the low forcasting
value in terms of efficiency, as shown by the graph
representation.
5. CONCLUSIONS
We investigated various load balancing strategies in the
Cloud Computing contextinthisarticle.We'vealsogoneover
some of the primary difficulties that must be considered
while developing any load balancing algorithm. There is no
concept of a one-time password in existing work that is
based on single authentication, where only the user name
Technique
Approach Date
packets
Proposed Existing
1024 3374ms 3889ms
2048 4098ms 4158ms
3072 5229ms 5344ms
4096 5310ms 5391ms
Figure 3 Flowchart of proposed algorithm

and password are required. To address this issue, a key-
based authentication system for multiple users has been
implemented to safeguard data. A one-time password is
issued and expires once the procedure is completed. A
emphasis will also be placed on throughput, with many
processes running at the same time. Using the proposed
approach, we were able to overcome the drawbacks of
earlier algorithms. A comparisonwasmade basedonseveral
factors such as throughput, reaction time,andsoon.Existing
algorithms are used to calculate the performance of various
topologies based on computing time and throughput.
6. FUTURE WORK
More cloud Computation over the Environment providers,
unlimited services, established quality, and best practises
can be expected in the future. In the future, we'll concentrate
on developing algorithms that maintain a better trade-off
between all performance characteristics.
REFERENCES
[1]. Xun Shao *, Masahiro Jibiki, Yuuichi Teranishi, Nozomu
Nishinaga, A proficient burden adjusting component for
heterogeneous reach queriable cloud storage✩,0167-
739X/© 2017
[2].Tinghuai Ma, Ya Chu, Licheng Zhao and Otgonbayar
Ankhbayar, Resource Allocation and Scheduling in Cloud
Computing: Policy and AlgorithmIETE Technical survey
Volume 31, Issue 1, January 2018, pages 4-16
[3]. B. P. Rima, E. Choi, and I. Lumb, A Taxonomy and Survey
of Cloud Computing Systems, Proceedings of fifth IEEE
International Joint Conference on INC, IMS and IDC, Seoul,
Korea, August 2017, pages 44-5l.
[4].Nidhi Jain Kansal, InderveerChana,CloudLoadBalancing
Techniques: A Step Towards Green Computing, IJCSI, Vol. 9,
Issue 1, January 2017.
[5].Hitesh Bheda, HirenBhatt,AnOverviewofLoadadjusting
Techniques in Cloud Computing Environments,Research
entryway JANUARY 2017.
[6].Hajara Idris., An improved insect state enhancement
calculation with adaptation to non-critical failure for work
planning for lattice registering frameworks, May 17, 2016
[7].Sukhvir Kaur, Supriya Kinger Review on Load Balancing
Techniques in Cloud Computing Environment, International
Journal of Science and Research (IJSR) 2017.
[8].X. Xu, W. Dou, X. Zhang, and J. Chen, Enreal: an
energyaware asset distribution strategy for logical work
process executions in cloud climate, 2016.
[9].Dr. Amit Agarwal, Saloni Jain Efficient ideal calculationof
assignment planning for distributed computing climate
International Journal of PC Trends and Technology
(IJCTT).2017
[10].X. Xu, W. Dou, X. Zhang, and J. Chen, Enreal: an
energyaware asset designation strategy for logical work
process executions in cloud climate, 2017.
[11].J. M. Galloway, K. L. Smith, and S. S. Vrbsky, Power
mindful burden adjusting for distributed computing, in
Proceedings of the World Congress on Engineering and
Computer Science, vol. 1, pp. 19–21, 2018.
[12].S. Sethi, A. Sahu, and S. K. Jena, Efficient burden
adjusting in distributed computing utilizing fluffy logic,IOSR
Journal of Engineering, vol. 2, no. 7, pp. 65–71, 2016.
[13]. T. Kokilavani J.J. School of Engineering and Technology
and Research Scholar, Bharathiar University, Tamil Nadu,
India, LoadBalancedMin-MinAlgorithmforStaticMeta-Task
Scheduling in Grid ComputingInternational Journal of
Computer Applications(0975 – 8887)Volume20–No.2,April
2017.
[14].M. Akbari and H. Rashidi, A multi-destinations booking
calculation dependent on cuckoo streamlining for task
allotment issue at arrange time in heterogeneous
frameworks, Expert Systems with Applications, vol. 60, pp.
234–248, 2018.
[15].Zenon Chaczko, Venkatesh Mahadevan, Shahrzad
Aslanzadeh, Christopher Mcdermid (2011),Availabity and
Load Balancing in Cloud Computing International
Conference on Computer and Software Modeling IPCSIT
vol.14 IACSIT Press,Singapore 2015.

Cloud Computing Task Scheduling Algorithm Based on Modified Genetic Algorithm

Recommended

More Related Content

Similar to Cloud Computing Task Scheduling Algorithm Based on Modified Genetic Algorithm (20)

More from IRJET Journal (20)

Recently uploaded (20)

Cloud Computing Task Scheduling Algorithm Based on Modified Genetic Algorithm