![International Journal of Trend in Scientific Research and Development (IJTSRD)
Volume 3 Issue 5, August 2019 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470
@ IJTSRD | Unique Paper ID – IJTSRD26361 | Volume – 3 | Issue – 5 | July - August 2019 Page 453
A Review on Novel Approach for
Load Balancing in Cloud Computing
Sukhdeep Kaur, Preeti Sondhi
UIET, Lalru, Punjab, India
How to cite this paper: Sukhdeep Kaur |
Preeti Sondhi "A Review on Novel
Approach for Load Balancing in Cloud
Computing"
Published in
International
Journal of Trend in
Scientific Research
and Development
(ijtsrd), ISSN: 2456-
6470, Volume-3 |
Issue-5, August 2019, pp.453-455,
https://doi.org/10.31142/ijtsrd26361
Copyright © 2019 by author(s) and
International Journalof Trendin Scientific
Research and Development Journal. This
is an Open Access article distributed
under the terms of
the Creative
CommonsAttribution
License (CC BY 4.0)
(http://creativecommons.org/licenses/by
/4.0)
ABSTRACT
Cloud computing is an interconnection between the networks such as in
private or public networks through internet in order to provide access to the
application, data and file storage. It not only decreases thecomputational cost,
hosting application, content storage and delivery rate. It is also a practical
approach in which data centre is transferred from a capital-intensivesetupto
a variable priced environment. As compared to traditional concepts, cloud
computing coveys the concept of the grid computing, distributed computing,
utility computing or autonomic computing. When any virtual machine gets
overloaded, fault may occur in the cloud environment. With the help of BFO
algorithm, technique of adaptive task scheduling is proposed. Using this
method, it becomes easy to transfer the task to the most reliable virtual
machine. In this research work, the technique will be proposed which will
select the most reliable virtual machine for the load balancing. The proposed
improvement leads to reduce execution time and resource consumption.
KEYWORDS: Weight-based algorithm, Load Balancing, CloudSim
1. INTRODUCTION
The cloud provides the feasibility to the user so that it can access the
information from anywhere. Therefore it removes the issue of location
constrained as in the traditional computers a set up was required to access the
information that is placed in other data storage device.
With the help of cloud physical relocation of human is
minimized as it gets access its data storage from anywhere.
As if somebody is using Cloud computing then their
resources get shared along with that the cost is also getting
shared. This helps user to spend only less cost as they have
to pay on the basis of usage [1]. The applications are
implemented on Public, Private or Hybrid clouds.Inorderto
determine the impacts cloud integrators has been utilized
and right path for the cloud for each organization.Despite its
developing influence, concerns regarding cloud computing
still remain. The benefits exceed the drawbacks and the
model merits exploring [2]. Some common challenges are:
Data Protection, Data Recovery and Availability, it includes
Appropriate clustering and Fail over, Data Replication,
System monitoring (Transactions monitoring, logs
monitoring and others),Maintenance(RuntimeGovernance)
and many more, Management Capabilities The management
of platform and infrastructure is not easy as there are
multiple cloud providers [3]. For some enterprises, features
like Auto-scaling are considered as the crucial requirement.
It has the great potential that it increases the scalability and
load balancing features, Regulatory and Compliance
Restrictions There are some countries that do not permit
customer's personal information and other sensitive
information to be leaked outside the state or country. To
resolve this issue cloud providers have to made data centre
or storage site within the country and it is not feasible to
provide such a infrastructure separately.
Cloud computing gives various computing archetype to the
numerous project, clients and online organizations, as the
resources can be utilized on demand. The main objective of
the cloud resource suppliers and consumers isthe allocation
of the cloud resources and accomplish the financial profit.
The major issue in the cloud computing is the allocation of
the resources as they are rarely distributed.
To Overcome all these challenges in cloud various
techniques have been proposed from time to time. The
different strategies are below:
Bio-Inspired Techniques
For the purpose of the search, there are various areas that
are required such as connectionism, socialbehaviouraswell
as emergence. In these techniques, the area of biology,
computer science as well as mathematics is covered. For the
biologically inspired computing techniques these
computational methods are considered as a broader view
that provides various applications [4]. In order to study the
IT oriented paradigm of cell computation or information
processing, various genetic algorithms and evolutionary
algorithms are proposed by many researchers. This
technique is considered as a powerfultool.Forthe validation
of the theories of biological evolutions and natural
inspirations, various algorithms are proposed using
mathematical optimization. This technique is not very
efficient as it is applicable at the small level and cannot be
utilized for the larger projects [5]. The main disadvantage is
misplaced of prominent genotype-phenotype idea of the
evolutionary algorithm.
IJTSRD26361](https://image.slidesharecdn.com/80areviewonnovelapproachforloadbalancingincloudcomputing-190907073012/85/A-Review-on-Novel-Approach-for-Load-Balancing-in-Cloud-Computing-1-320.jpg)
![International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470
@ IJTSRD | Unique Paper ID – IJTSRD26361 | Volume – 3 | Issue – 5 | July - August 2019 Page 454
Artificial Bee Colony Optimization
The artificial bee colony algorithm is an optimization
algorithm based on the meta-heuristics in which various
optimal numerical solution has been find out among a
substantial number of alternatives. This process is followed
while trying to tackle NP difficult problems. In this ABC
includes the three groups of the bee in the colony of the
artificial bee such as employed bees, onlookers and scouts.
Onlookers is referred as the behaviour,inwhich beetookthe
decision for a food source. When it goes to previouslyvisited
place it is named as employed bee. The random search
carried out by the bee is referred as scouts. In the
optimization problem, the position of a food source
represents a possible solution and the quality of thesolution
is described by using nectar amount of a food source. The
swarm of the bees is moved in the random direction in the
two-dimensional search space. When a nectar target is
discovered bees start interacting with each other and find
out the optimal solution for the problems obtained from the
intensity of these bee interactions.
In this algorithm, potential food source are initializedforthe
vector of the population. Initially, a new food source is
searched by the bees randomly.Aftertheidentificationofthe
food source, the fitness of the obtained food is identified and
calculated. Further if another food source is discovered by
the employed bees having a greater fitness, they utilized the
new sources and deserted the existing one. The fitness
information is transferred to the onlooker bees by the
employed bees as food selection by the onlooker bees done
on the basis of the probability of thefood occurring[6].Their
solutions are rejected, if the fitness of the food source is not
proved by the employed bees.
Intelligent Water Drops Algorithm (IWD)
Intelligent Water Drops Algorithm is proposed by Hamed
Shah-hosseini that is mainly based on the population based
strategy. The major inspiration to this system is the process
of Natural River systems in which actions and reactions are
occur between water drops in the river. It also follows the
changes in the environment such as river is flowing [7]. The
main logic behind this method is the use of the behavior of
the water drops, therefore, an artificial water drop is
developed by the researcher thatpossesssomeproperties of
the natural water drop. This Intelligent Water Drop has two
important properties:
The amount of the soil it carries now, Soil (IWD).
The velocity that it is moving now, Velocity (IWD).
The main problem is the environment from which the water
flows, hence it is important to take care of the environment
for this purpose.
2. Literature Review
Ji Su Park [1]: With the fast spread of mobile devices, a huge
amount data is generated in a mobile environment. The
distributed processing technologies such as MapReduceare
applied to mobiledevices,thankstotheimproved computing
power of mobile devices. However, mobile devices have
several problems such as the movement problem and the
utilization problem. Especially, the utilization problem and
the movement problem of mobile devices cause system
faults more frequently because of dynamic changes, and
system faults preventapplicationsusingmobiledevicesfrom
being processed reliably. In this author proposed scheme,
mobile devices are separated into groups by cut-off points
based on entropy values. He also proposes a two-phase
grouping method in order to reduce the overhead of group
management. The experimental result shows that our
algorithm outperforms traditionalgroupingtechniqueswith
maintaining stable big data processing and managing
reliable resource.
Liang Q [2]: In this work author puts forward a
reconfiguration framework based on a request prediction,
which anticipates theapplicationrequest volumein advance.
To determine the objective of relatively optimal
configuration, it can work out the allocation scheme which
can improve the resource utilization ratio as well as lower
energy consumption. In addition, a concept of Utility Ratio
Matrix (URM) is put forward torepresentallocationsof hosts
and Virtual Machines (VMs),andareconfiguration algorithm
based on request prediction is alsopresented. The algorithm
will predict the application requests so as to work out the
allocation scheme in advance. The algorithm can separate
the reconfiguration computing from the real allocation so
that it can avoid a time delay between the reconfiguration
result and the varied demands, and can also reduce the
energy consumption in data centre. The corresponding
analysis and experimental results indicate the feasibility of
the reconfiguration algorithm in this paper.
Christian V [3]: As scientific application require large
computing power, traditionally exceedingtheamount thatis
available within the premises of a single institution. In this
author developed Aneka’s deadline- driven provisioning
mechanism, which is responsible for supporting quality of
service (QoS)-aware execution of scientific applications in
hybrid clouds composed of resources obtained from a
variety of sources. Experimental results evaluating such a
mechanism show that Aneka is able to efficiently allocate
resources from different sources in order to reduce
application execution times
3. Proposed Work
In the cloud computing technology, task scheduling policy is
considered as a crucial component that provides the Quality
of Service to the whole cloud computing systems. Atrade-off
between user requirements and resource utilization is done
with the help of task scheduling strategy. In order toallocate
work, scheduling is considered as the resource method. The
main objective of this research work is
To study and analyzed the existing series of strategiesbased
on multi-objective task scheduling and to understand their
limitations.
To design and implement improved optimization algorithm
based on Bacterial foraging technique for task scheduling.
To increase the speed of processing by exploitation full
resources and to achieve high correctness and improved
identification rate.
To validate the proposed algorithm
In order to resolve the node failure within the cloud
networks, the BFO algorithm is proposed in this research
work. There are number of nodes available in the present
algorithm. On the basis of failure rate and least execution
time utilized, the candidate node will be selected form these
nodes. A threshold value is set here by the master node in](https://image.slidesharecdn.com/80areviewonnovelapproachforloadbalancingincloudcomputing-190907073012/85/A-Review-on-Novel-Approach-for-Load-Balancing-in-Cloud-Computing-2-320.jpg)
![International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470
@ IJTSRD | Unique Paper ID – IJTSRD26361 | Volume – 3 | Issue – 5 | July - August 2019 Page 455
which there are two parameters to be considered which are
failure rate and execution time. The master node selects the
candidate node on the basis of the node that hasequaltoand
less failure rate as well as minimum execution time. . Once
the candidate node is chosen, the execution of tasks will be
initiated. The numbers of tasks to be executed within this
scenario are also needed to be entered here. Failure will
occur at the point where one task moves from its location
during the execution of task. a novel technique is proposed
in this paper in order to solve this issue such that the
mobility of a node does not cause any kinds of failure within
the networks.
4. Conclusion and Future Work
There is a need to increase the number of data centers
according to the needs of host in order to ensure the Quality
of Service within the network. There will be an increment in
the energy being consumed by the network at fixed rate as
the number of data centres increase. Thus, the QoS can be
ensured at the required level through this. In this research
work, the BFO technique will be optimized for the virtual
machine migration. The virtual machine which is maximum
reliable on that machine task will bemigrated.Theproposed
improvement leads to reduce execution time and resource
consumption
.
5. References
[1] JiSu P, Hyongsoon K, Young-Sik J, Eunyoung L, 2014.
“Two-phase grouping-based resourcemanagementfor
big data processing in mobile cloud computing”, Int J
Commun Syst volume27, issue 3, pp-839–851.
[2] Liang Q, Zhang J, Zhang YH, Liang JM, 2014. “The
placement method of resources and applicationsbased
on request prediction in cloud data center”, Inf Sci,
volume 15, issue 9, pp-735–745.
[3] J. H. Holland, 1973. “Geneticalgorithms andtheoptimal
allocation of trials”, SIAM J. Comput. Volume 2, issue 2,
pp. 88–105
[4] Bonabeau, E., Dorigo, M. and Theraulaz, G.1999.
“Swarm intelligence.Oxford UniversityPress”, research
publications, volume 14, issue 5, pp- 184-193.
[5] Kennedy, J.; Eberhart, R. (1995). “Particle Swarm
Optimization”, Proceedings of IEEE International
Conference on Neural Networks, volume 4, issue 6, pp.
1942–1948.
[6] Christian V, Rodrigo NC, Dileban K, Rajkumar B, 2012.
“Deadline-driven provisioning of resources for
scientific applications in hybrid clouds with Aneka”,
Future Gener Comput Syst volume28, issue16,pp-58–
65
[7] Bowman-Amuah, M. K. 2003. “Load balancer in
environment services patterns”, U.S. Patent,
Washington, DC: U.S. Patent and Trademark Office,
volume 3, issue 4, pp- 239-246.
[8] Koza, John R. 1992 “Genetic Programming: On the
Programming of Computers by Means of Natural
Selection”, Cambridge, MA: The MIT Press. Volume 9,
issue 13, pp- 118-124
[9] Beyer, H.G. and Schwefel, H.P. 2002. “Evolution
strategies. NaturalComputingvolume1,issue3, pp-52-
59.
[10] R. Storn, K. Price, 1997. “Differential evolution – a
simple and efficient heuristic for global optimization
over continuous spaces, Journal of GlobalOptimization
volume 11, issue 3, pp- 341–359.](https://image.slidesharecdn.com/80areviewonnovelapproachforloadbalancingincloudcomputing-190907073012/85/A-Review-on-Novel-Approach-for-Load-Balancing-in-Cloud-Computing-3-320.jpg)
A Review on Novel Approach for Load Balancing in Cloud Computing
- 1. International Journal of Trend in Scientific Research and Development (IJTSRD) Volume 3 Issue 5, August 2019 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470 @ IJTSRD | Unique Paper ID – IJTSRD26361 | Volume – 3 | Issue – 5 | July - August 2019 Page 453 A Review on Novel Approach for Load Balancing in Cloud Computing Sukhdeep Kaur, Preeti Sondhi UIET, Lalru, Punjab, India How to cite this paper: Sukhdeep Kaur | Preeti Sondhi "A Review on Novel Approach for Load Balancing in Cloud Computing" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456- 6470, Volume-3 | Issue-5, August 2019, pp.453-455, https://doi.org/10.31142/ijtsrd26361 Copyright © 2019 by author(s) and International Journalof Trendin Scientific Research and Development Journal. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (CC BY 4.0) (http://creativecommons.org/licenses/by /4.0) ABSTRACT Cloud computing is an interconnection between the networks such as in private or public networks through internet in order to provide access to the application, data and file storage. It not only decreases thecomputational cost, hosting application, content storage and delivery rate. It is also a practical approach in which data centre is transferred from a capital-intensivesetupto a variable priced environment. As compared to traditional concepts, cloud computing coveys the concept of the grid computing, distributed computing, utility computing or autonomic computing. When any virtual machine gets overloaded, fault may occur in the cloud environment. With the help of BFO algorithm, technique of adaptive task scheduling is proposed. Using this method, it becomes easy to transfer the task to the most reliable virtual machine. In this research work, the technique will be proposed which will select the most reliable virtual machine for the load balancing. The proposed improvement leads to reduce execution time and resource consumption. KEYWORDS: Weight-based algorithm, Load Balancing, CloudSim 1. INTRODUCTION The cloud provides the feasibility to the user so that it can access the information from anywhere. Therefore it removes the issue of location constrained as in the traditional computers a set up was required to access the information that is placed in other data storage device. With the help of cloud physical relocation of human is minimized as it gets access its data storage from anywhere. As if somebody is using Cloud computing then their resources get shared along with that the cost is also getting shared. This helps user to spend only less cost as they have to pay on the basis of usage [1]. The applications are implemented on Public, Private or Hybrid clouds.Inorderto determine the impacts cloud integrators has been utilized and right path for the cloud for each organization.Despite its developing influence, concerns regarding cloud computing still remain. The benefits exceed the drawbacks and the model merits exploring [2]. Some common challenges are: Data Protection, Data Recovery and Availability, it includes Appropriate clustering and Fail over, Data Replication, System monitoring (Transactions monitoring, logs monitoring and others),Maintenance(RuntimeGovernance) and many more, Management Capabilities The management of platform and infrastructure is not easy as there are multiple cloud providers [3]. For some enterprises, features like Auto-scaling are considered as the crucial requirement. It has the great potential that it increases the scalability and load balancing features, Regulatory and Compliance Restrictions There are some countries that do not permit customer's personal information and other sensitive information to be leaked outside the state or country. To resolve this issue cloud providers have to made data centre or storage site within the country and it is not feasible to provide such a infrastructure separately. Cloud computing gives various computing archetype to the numerous project, clients and online organizations, as the resources can be utilized on demand. The main objective of the cloud resource suppliers and consumers isthe allocation of the cloud resources and accomplish the financial profit. The major issue in the cloud computing is the allocation of the resources as they are rarely distributed. To Overcome all these challenges in cloud various techniques have been proposed from time to time. The different strategies are below: Bio-Inspired Techniques For the purpose of the search, there are various areas that are required such as connectionism, socialbehaviouraswell as emergence. In these techniques, the area of biology, computer science as well as mathematics is covered. For the biologically inspired computing techniques these computational methods are considered as a broader view that provides various applications [4]. In order to study the IT oriented paradigm of cell computation or information processing, various genetic algorithms and evolutionary algorithms are proposed by many researchers. This technique is considered as a powerfultool.Forthe validation of the theories of biological evolutions and natural inspirations, various algorithms are proposed using mathematical optimization. This technique is not very efficient as it is applicable at the small level and cannot be utilized for the larger projects [5]. The main disadvantage is misplaced of prominent genotype-phenotype idea of the evolutionary algorithm. IJTSRD26361
- 2. International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470 @ IJTSRD | Unique Paper ID – IJTSRD26361 | Volume – 3 | Issue – 5 | July - August 2019 Page 454 Artificial Bee Colony Optimization The artificial bee colony algorithm is an optimization algorithm based on the meta-heuristics in which various optimal numerical solution has been find out among a substantial number of alternatives. This process is followed while trying to tackle NP difficult problems. In this ABC includes the three groups of the bee in the colony of the artificial bee such as employed bees, onlookers and scouts. Onlookers is referred as the behaviour,inwhich beetookthe decision for a food source. When it goes to previouslyvisited place it is named as employed bee. The random search carried out by the bee is referred as scouts. In the optimization problem, the position of a food source represents a possible solution and the quality of thesolution is described by using nectar amount of a food source. The swarm of the bees is moved in the random direction in the two-dimensional search space. When a nectar target is discovered bees start interacting with each other and find out the optimal solution for the problems obtained from the intensity of these bee interactions. In this algorithm, potential food source are initializedforthe vector of the population. Initially, a new food source is searched by the bees randomly.Aftertheidentificationofthe food source, the fitness of the obtained food is identified and calculated. Further if another food source is discovered by the employed bees having a greater fitness, they utilized the new sources and deserted the existing one. The fitness information is transferred to the onlooker bees by the employed bees as food selection by the onlooker bees done on the basis of the probability of thefood occurring[6].Their solutions are rejected, if the fitness of the food source is not proved by the employed bees. Intelligent Water Drops Algorithm (IWD) Intelligent Water Drops Algorithm is proposed by Hamed Shah-hosseini that is mainly based on the population based strategy. The major inspiration to this system is the process of Natural River systems in which actions and reactions are occur between water drops in the river. It also follows the changes in the environment such as river is flowing [7]. The main logic behind this method is the use of the behavior of the water drops, therefore, an artificial water drop is developed by the researcher thatpossesssomeproperties of the natural water drop. This Intelligent Water Drop has two important properties: The amount of the soil it carries now, Soil (IWD). The velocity that it is moving now, Velocity (IWD). The main problem is the environment from which the water flows, hence it is important to take care of the environment for this purpose. 2. Literature Review Ji Su Park [1]: With the fast spread of mobile devices, a huge amount data is generated in a mobile environment. The distributed processing technologies such as MapReduceare applied to mobiledevices,thankstotheimproved computing power of mobile devices. However, mobile devices have several problems such as the movement problem and the utilization problem. Especially, the utilization problem and the movement problem of mobile devices cause system faults more frequently because of dynamic changes, and system faults preventapplicationsusingmobiledevicesfrom being processed reliably. In this author proposed scheme, mobile devices are separated into groups by cut-off points based on entropy values. He also proposes a two-phase grouping method in order to reduce the overhead of group management. The experimental result shows that our algorithm outperforms traditionalgroupingtechniqueswith maintaining stable big data processing and managing reliable resource. Liang Q [2]: In this work author puts forward a reconfiguration framework based on a request prediction, which anticipates theapplicationrequest volumein advance. To determine the objective of relatively optimal configuration, it can work out the allocation scheme which can improve the resource utilization ratio as well as lower energy consumption. In addition, a concept of Utility Ratio Matrix (URM) is put forward torepresentallocationsof hosts and Virtual Machines (VMs),andareconfiguration algorithm based on request prediction is alsopresented. The algorithm will predict the application requests so as to work out the allocation scheme in advance. The algorithm can separate the reconfiguration computing from the real allocation so that it can avoid a time delay between the reconfiguration result and the varied demands, and can also reduce the energy consumption in data centre. The corresponding analysis and experimental results indicate the feasibility of the reconfiguration algorithm in this paper. Christian V [3]: As scientific application require large computing power, traditionally exceedingtheamount thatis available within the premises of a single institution. In this author developed Aneka’s deadline- driven provisioning mechanism, which is responsible for supporting quality of service (QoS)-aware execution of scientific applications in hybrid clouds composed of resources obtained from a variety of sources. Experimental results evaluating such a mechanism show that Aneka is able to efficiently allocate resources from different sources in order to reduce application execution times 3. Proposed Work In the cloud computing technology, task scheduling policy is considered as a crucial component that provides the Quality of Service to the whole cloud computing systems. Atrade-off between user requirements and resource utilization is done with the help of task scheduling strategy. In order toallocate work, scheduling is considered as the resource method. The main objective of this research work is To study and analyzed the existing series of strategiesbased on multi-objective task scheduling and to understand their limitations. To design and implement improved optimization algorithm based on Bacterial foraging technique for task scheduling. To increase the speed of processing by exploitation full resources and to achieve high correctness and improved identification rate. To validate the proposed algorithm In order to resolve the node failure within the cloud networks, the BFO algorithm is proposed in this research work. There are number of nodes available in the present algorithm. On the basis of failure rate and least execution time utilized, the candidate node will be selected form these nodes. A threshold value is set here by the master node in
- 3. International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470 @ IJTSRD | Unique Paper ID – IJTSRD26361 | Volume – 3 | Issue – 5 | July - August 2019 Page 455 which there are two parameters to be considered which are failure rate and execution time. The master node selects the candidate node on the basis of the node that hasequaltoand less failure rate as well as minimum execution time. . Once the candidate node is chosen, the execution of tasks will be initiated. The numbers of tasks to be executed within this scenario are also needed to be entered here. Failure will occur at the point where one task moves from its location during the execution of task. a novel technique is proposed in this paper in order to solve this issue such that the mobility of a node does not cause any kinds of failure within the networks. 4. Conclusion and Future Work There is a need to increase the number of data centers according to the needs of host in order to ensure the Quality of Service within the network. There will be an increment in the energy being consumed by the network at fixed rate as the number of data centres increase. Thus, the QoS can be ensured at the required level through this. In this research work, the BFO technique will be optimized for the virtual machine migration. The virtual machine which is maximum reliable on that machine task will bemigrated.Theproposed improvement leads to reduce execution time and resource consumption . 5. References [1] JiSu P, Hyongsoon K, Young-Sik J, Eunyoung L, 2014. “Two-phase grouping-based resourcemanagementfor big data processing in mobile cloud computing”, Int J Commun Syst volume27, issue 3, pp-839–851. [2] Liang Q, Zhang J, Zhang YH, Liang JM, 2014. “The placement method of resources and applicationsbased on request prediction in cloud data center”, Inf Sci, volume 15, issue 9, pp-735–745. [3] J. H. Holland, 1973. “Geneticalgorithms andtheoptimal allocation of trials”, SIAM J. Comput. Volume 2, issue 2, pp. 88–105 [4] Bonabeau, E., Dorigo, M. and Theraulaz, G.1999. “Swarm intelligence.Oxford UniversityPress”, research publications, volume 14, issue 5, pp- 184-193. [5] Kennedy, J.; Eberhart, R. (1995). “Particle Swarm Optimization”, Proceedings of IEEE International Conference on Neural Networks, volume 4, issue 6, pp. 1942–1948. [6] Christian V, Rodrigo NC, Dileban K, Rajkumar B, 2012. “Deadline-driven provisioning of resources for scientific applications in hybrid clouds with Aneka”, Future Gener Comput Syst volume28, issue16,pp-58– 65 [7] Bowman-Amuah, M. K. 2003. “Load balancer in environment services patterns”, U.S. Patent, Washington, DC: U.S. Patent and Trademark Office, volume 3, issue 4, pp- 239-246. [8] Koza, John R. 1992 “Genetic Programming: On the Programming of Computers by Means of Natural Selection”, Cambridge, MA: The MIT Press. Volume 9, issue 13, pp- 118-124 [9] Beyer, H.G. and Schwefel, H.P. 2002. “Evolution strategies. NaturalComputingvolume1,issue3, pp-52- 59. [10] R. Storn, K. Price, 1997. “Differential evolution – a simple and efficient heuristic for global optimization over continuous spaces, Journal of GlobalOptimization volume 11, issue 3, pp- 341–359.