ANALYSIS ON LOAD BALANCING ALGORITHMS IMPLEMENTATION ON CLOUD COMPUTING ENVIRONMENT

International Journal of Innovative Research in Advanced Engineering (IJIRAE) ISSN: 2349-2163
Issue 02, Volume 6 (February 2019) www.ijirae.com
___________________________________________________________________________________________________
IJIRAE: Impact Factor Value – Mendeley (Elsevier Indexed); Citefactor 1.9 (2017); SJIF: Innospace, Morocco (2016):
3.916 | PIF: 2.469 | Jour Info: 4.085 | ISRAJIF (2017): 4.011 | Indexcopernicus: (ICV 2016): 64.35
IJIRAE © 2014- 19, All Rights Reserved Page–32
ANALYSIS ON LOAD BALANCING ALGORITHMS
IMPLEMENTATION ON CLOUD COMPUTING
ENVIRONMENT
Dina Farouk Altayeb
Department of Computer Engineering, Faculty of Engineering,
Future University, Egypt
dina-farouk@hotmail.com;
Fatima Abdelghani Mustafa
Department of Computer Engineering, Faculty of Engineering,
Future University, Egypt
Manuscript History
Number: IJIRAE/RS/Vol.06/Issue02/FBAE10081
Received: 02, February 2019
Final Correction: 09, February 2019
Final Accepted: 14, February 2019
Published: February 2019
Citation: Dina & Fatima (2019). Analysis on Load Balancing Algorithms Implementation on Cloud Computing
Environment. IJIRAE::International Journal of Innovative Research in Advanced Engineering, Volume VI, 32-36.
doi://10.26562/IJIRAE.2019.FBAE10081
Editor: Dr.A.Arul L.S, Chief Editor, IJIRAE, AM Publications, India
Copyright: ©2019 This is an open access article distributed under the terms of the Creative Commons Attribution
License, Which Permits unrestricted use, distribution, and reproduction in any medium, provided the original
author and source are credited
Abstract- Cloud computing means storing and accessing data and programs over the Internet instead of your
computer's hard drive. The cloud is just a metaphor for the Internet. The elements involved in cloud computing are
clients, datacenter and distributed server. One of the main problems in cloud computing is load balancing. Balancing
the load means to distribute the workload among several nodes evenly so that no single node will be overloaded.
Load can be of any type that is it can be CPU load, memory capacity or network load. In this paper we presented an
architecture of load balancing and algorithm which will further improve the load balancing problem by minimizing
the response time. In this paper, we have proposed the enhanced version of existing regulated load balancing
approach for cloud computing by comping the Randomization and greedy load balancing algorithm. To check the
performance of proposed approach, we have used the cloud analyst simulator (CloudAnalyst). Through simulation
analysis, it has been found that proposed improved version of regulated load balancing approach has shown better
performance in terms of cost, response time and data processing time.
Keywords: Cloud computing; Load Balancing; Data Center; Virtual Machines; Cloud Analyst;
I. INTRODUCTION
Cloud Computing can be considered as a platform for development, maintenance and accessing applications by user
by paying for the resources which are only used for certain time. To serve for huge number of requests from
different types of users located at different parts of the world on pay-per-usage bases, the process of virtualization
has been followed in cloud computing environment. There are so many service providers who are responsible for
maintaining the application on cloud environment. The most predominant cloud service providers are Google,
Amazon, Microsoft and many others. In order to serve the huge traffic to world these service providers maintain
data centers all over the world where the data is stored in bulk and requests are processed [4]. Depending on the
number of requests to be processed by a data center, no. of virtual machines is created where on a single CPU
different operating systems and their configurations can be run. It gives the illusion to the users that, there are
numbers of CPUs are involved for processing the requests. This whole process can be considered as virtualization
and it is monitored by the software called as hypervisor. The hypervisor is mainly responsible in creating and
maintaining the virtual machines. The requests from all the users are considered as individual request and each of
the virtual machines is assigned a request and effort is made to keep them busy for longer time.

___________________________________________________________________________________________________
This is considered to be load balancing policy which makes an effort to maximize the throughput of virtual
machines. Job Scheduling is a process of allocating jobs onto available resources in time. It is also defined as the
process of finding an efficient mapping of tasks to the suitable resources so that the execution can be completed
with the satisfaction of some objective functions. In short, more efficient is the scheduling algorithm, better is the
Quality of Service delivered. Every Scheduling problem has three important elements [3]. They are:
Machine Configuration: A single machine with a single or multiple processors or a cluster of machines with a
single or multiple processors in each machine etc.
Optimization Criterion: It defines the objective(s) of the scheduling algorithm e.g. reducing make span, minimizing
response time, minimizing resource cost etc. The main intention for developing this tool is to simulate the traffic
generated by most visited applications such as face book, Gmail and analyze the response times at each data center.
The simulation process is divided into regions denoted by Ri where ‘i’ indicates the region number and number of
requests generated in each region can be considered as a User Base denoted by UBi where ‘i’ indicates the user base
number. The data centers are denoted by DCi where ‘i’ indicates the corresponding data center number. The
following figure gives a better understanding of how the regions are divided for simulation and the table1
summarizes the regions. Elaborates each and every detail of the tool[9].
II. LOAD BALACING
Load balancing is defining as the distribution of resources, simultaneous working of the schedulers, efficiency
enhancement, and minimization of response time via a suitable matching of job to the available resource.
Simultaneous working of the schedulers involves the distribution of load in equal manner among the processors. To
restore the balance dynamic load balancing also known as load sharing or load migration is employed [4]. It is done
by distributing the entire load to the individual processors of the complete structure for obtaining efficient resource
mapping and concurrently removing the possibility of overloading or under loading of the nodes in the network. It is
done to achieve for better ratio of user realization and resource utilization, thereby enhancing the throughput of the
complete system. If done in proper manner the load management can limit the consumption of the available. It also
helps in executing failures, making the system scalable, and over-burdening, minimizing response time etc.
III. RELATED WORK
Load balancing is used to distributing the load across multiple nodes for enhancing the overall performance of a
system. The current load balancing algorithms in cloud computing environment is not highly efficient. Load
balancing is very complex task today, because the users request arrival on server is not possible i.e., we cannot
predict it. Each Vms [1] [4] has different specifications. So it is difficult to schedule the job and balance the nodes.
Recently, many research work done on load balancing. Load balancing mainly classified into two categories, static
load balancing and dynamic load balancing algorithms. Static load balancing algorithms mainly defined in the design
or implementation of the system. Dynamic load balancing algorithm considered only current state of the system
during load balancing. The existing algorithms are following: In paper [5] have described a conventional round
robin approach for balancing the load. A group of available VMs gets the tasks on the random basis and the process
of task allocation continues in circular (round) motion. When a task is mapped to the VM then it goes to the last
position in the VM list. The discussed approach doesn’t have any idea of size of the incoming tasks so suffers with
the disadvantage of some overloaded nodes. Besides this, the benefit of this algorithm is that inter-process
communication is not required. In paper [10] have presented a Weighted Round Robin approach for balancing the
load in cloud environment. The described scheme is a combination of weight assigning and round robin approach.
The capacity of the VM to accommodate the tasks helps in assigning weight to the VM and after selecting the VM
conventional round robin approach is executed. In paper [8] have discussed basic Throttled load balancing
approach for cloud environments in which it was considered that VM has the capacity to handle single task only and
the incoming tasks are assigned to the idle Vm’s which are selected randomly if more VMs are found to be idle. in
paper [10] have presented a technique to balance the load among the overloaded and under-loaded nodes by simply
shifting the jobs from overloaded node to under-loaded node in case any virtual machine is found overloaded. To
accomplish this task, the technique tracks the data of VM id, task id and no of active tasks allocated to the VM.
IV. PROBLEM IDENTIFICATION
Cloud computing is a term, which involves virtualization, distributed computing, networking and software and web
services”. As we talk about a cloud it consists many parameters like shoppers, datacenter & distributed system.
Cloud comprises of fault tolerance, convenience, and quantifiability, litheness, compact overhead for users, compact
value of possession etc. [2]. Load balancing is therefore may be defined as the method of allocate the load among
different nodes of a Data Center to enhance each resource employment and process latency whereas additionally
avoiding a state of affairs wherever a number of nodes are highly loaded whereas alternative nodes are idle or doing
little or no work, and some of the physical machines and virtual machines are having maximum imbalance level of
Cloud data centers [5].

___________________________________________________________________________________________________
V. SOLUTION DOMAIN
We proposed new hybrid load balancing algorithm, which is combination of randomizing and greedy load balancing
algorithm. Our aim is to improve the response time for the user (UserBase) and processing time of data center. Our
proposed Hybrid Algorithm by effective reallocation the tasks, it had deployment at the VmLoadBalancer in
Datacenter Controller, to distribute load among nodes (VM) are idle or doing little or no work. to improve overall
system response time.
Proposed Algorithm:
new request Output: The VM id that selected to assign the load. 0.Initialize, Cl(0..i-1) ← 0 At start all VM’s have zero
allocation., K← 50, vmid ←-1 ,i← 0,currCount ← 0, minCount ← Max_Value, Temp ← -1;
1. Parses VM_List() to LoadBalancer:
2. For i← 0 to k //Select VM randomly
3. Temp ← random(VmStatesList())
4. VMid ← Temp
5. If vmid Exist in Cl(VMid) then
6. currCount ← ClTable(VMid), else
7. currCount ← 0
8. VMids() ← (VMid, currCount).
9. End for
10.Temp ← -1
11.currCount ← 0
12.For i ← 0 to k
13.TempVMid ← i
14. currCount ← VMids(TempVMid)
15. If currCount <minCount then
16. minCount= currCount
17. Vmid ← TempVMid
18. End if
19. End for
20. Cl(VMid) ← Cl(VMid+1)
21. return vmId;
The load balancer spreads the load on to completely different nodes, and hence, it's referred to as unfold spectrum
technique. The load balancer maintains a queue of the roles that require using and are presently mistreatment the
services of the virtual machine. The balancer then unendingly scans this queue and therefore the list of virtual
machines. We implemented hybrid algorithm, on Eclipse using advance java, and cloud Simulation CloudAnalyst. we
configure many parameters like number of datacenters, number of cloudlets, VM configuration, bandwidth and
MIPS. We implemented three algorithm of load balancing are:
 Throttled.
 Equally Spread Current Execution.
 Round Robin.
 Hybrid Algorithm.
V. RESULT SET
Cloud load balancing is developed in this research with help Java (JDK1.8) and Eclipse IDE8.02 on window operating
system 10. In Result Analysis Compare proposed system with existing system in term of Average Response time and
CPU Factor.
First step: This is the first page of our project, which is shown in figure1.
Figure 1: Cloud Analyst Simulator front page First Step

___________________________________________________________________________________________________
Second Step: Adding Hybrid Algorithm in CloudAnalyst.
Figure 2: Adding Hybrid Algorithm in CloudAnalyst.
Third Step: Configure Two Datacenter and Implement VMs.
Figure 3 Configure Datacenter and Implement VMs.
Fourth Step: run simulation for all Algorithm
Figure 4: Demonstration of proposed work in Third step.
Table 1: shows average, Minimum, and Maximum time for each Load Balancing algorithm used in this paper.
Hybrid Algorithm Round Robin Throttled ESCE
DC RT (Avg) 51.51 52.82 51.59 52.36
DC RT(Min) 34.37 34.93 34.39 34.75
DC RT(Max) 138.34 166.70 162.99 175.29
Fifth Step: we repeat all step above for all other load balancing algorithm and round robin, ESCE, Throttled)
Fifth Step: compare the results with other in term of Average time

___________________________________________________________________________________________________
Figure 5 Two Data Center Efficiency on Different Region (US-UK)
VI. CONCLUSION AND FUTURE WORK
Load balancing considered as one of the most challenges in cloud computing, it is the major factor to improve the
performance of the cloud computing. In This paper proved the Load balancing helps to distributing the total load to
the individual available VM to enhance resource utilization and response time. It also considered a situation where
some of the nodes are heavily loaded while other nodes are idle or doing very little work. Here, implemented the
new algorithms and compared the performance with the existing load balancing algorithm using the simulator
CloudAnalyst. The performance was compared based on Response time computing with respect to stability,
resource utilization, dynamicity. In Future either we discussed only on improving the performance on one data
center, but there are still other approaches that can be applied to balance the load in clouds computing
environment, improved algorithm or combination of algorithms will improve performance of cloud load balancing.
In Future also compare load-balancing algorithms on other parameters.
REFERENCES
1. Mohamed Riduan Abid, Moulay Idriss El Ouadghiri, Michael Gerndt “Virtual Machines’ Load-Balancing in Inter-
Clouds” 2016 4th International Conference on Future Internet of Things and Cloud Workshops , © 2016 IEEE
http://doiI://10.1109/W-FiCloud.2016.35
2. Rajkumar somani, Jyotsana Ojha, “A Hybrid approach for Vm load balancing in cloud using
cloudsim”,2014,International Journal of science, Engineering and Technology Research(IJSETR),Volume 3.Issue
6,June 2014
3. Ritu Kapur, “A Cost Effective approach for Resource Scheduling in Cloud Computing”, 2015, IEEE International
Conference on Computer, Communication and Control(IC4-2015).
4. Dr. Rakesh Rathi1, Vaishali Sharma2 and Sumit Kumar Bole3, “Round Robin Data Center Selection in Single
Region for Service Proximity Service Broker in Cloud Analyst”, International Journal of Computer & Technology,
Volume 4 no. 2, March- April 2013.
5. Bhatiya Wickremansinghe1, Rodrigo N. Calheiros2and Dr. Raj kumar Buyya3, “CloudAnalyst: A Cloud Sim- based
Visul Modeller for Analysing Cloud Computing Environments and Applications”, IEEE Computer Society, 2010.
6. Kunal Mahurkar1, Shraddha Katore2 and Suraj Bhaisade3, Pratikawale4, “Reducing Cost of Provisioning in Cloud
Computing”, International Journal of Advance in Computer Science and Cloud Computing, Volume- 1, Issue- 2,
nov.- 2013,
7. Syed Tauhid Zuheri1, Tamanna Shamrin2 and Rusia Tanbin3, Firoj Mahmud4, “An Efficient Load Balancing
Approach in Cloud Environment by using Round Robin Algorithm”, International Journal of Artificial and
Mechatronics, volume 1, issue 5, 2013.
8. Tiwari, M., Gautam, K., and Katare, K., Analysis of Public Cloud Load Balancing using Partitioning Method and
Game Theory. International Journal of Advanced Research in Computer Science and Software Engineering, 4(2):
pp. 807-812,(2014).
9. Sareen, P., Cloud Computing: Types, Architecture, Applications, Concerns, Virtualization and Role of IT
Governance in Cloud. International Journal of Advanced Research in Computer Science and Software
Engineering, 3(3): pp. 533-538,(2013).
10.Ray, S. and De Sarkar, A., Execution Analysis Of Load Balancing Algorithms In Cloud Computing Environment.
International Journal on Cloud Computing: Services and Architecture (IJCCSA), 2(5):pp. 1-13,(2012).

ANALYSIS ON LOAD BALANCING ALGORITHMS IMPLEMENTATION ON CLOUD COMPUTING ENVIRONMENT

Recommended

More Related Content

What's hot (20)

Similar to ANALYSIS ON LOAD BALANCING ALGORITHMS IMPLEMENTATION ON CLOUD COMPUTING ENVIRONMENT (20)

More from AM Publications (20)

Recently uploaded (20)

ANALYSIS ON LOAD BALANCING ALGORITHMS IMPLEMENTATION ON CLOUD COMPUTING ENVIRONMENT