TOWARDS EFFICIENT CLOUD SERVICES :

AN ENERGY –AWARE HYBRID LOAD

BALANCING APPROACH

Presented by- Guided by-

1. S.PAVITHRA (733119405001) Mr.M.Manikandan .M.E (PhD)
HOD
 ABSTRACT
• The explosive growth of cloud computing in present will lead to a massive increase in
both the amount of traffic and the number of service requests to cloud servers.

• This will cause overload on the server and lead to the degradation of the performance of
servers.

• Load balancing is the process of distributing load on various nodes providing best
resource utilization when nodes are overloaded. It includes optimal utilization of
resources which increases the performance of the system and thereby it minimizes the
load of competent resources.
• The main objective is to analyze the issues of existing load balancing algorithms and
proposes a Multi objective load balancing transformation technique for improving the
response time of VMs on cloud computing to improve performance for end-user and also
utilizes Cuckoo Search algorithm by considering the various QoS parameters including
service response time, reliability, cost, and estimate the power consumption of VMs on a
host to enhance load balancing.
• The Proposed algorithm improves the response times and processing time of the cloud
data center.
CONTENT
• INTRODUCTION
• LITERATURE SURVEY
• EXISTING SYSTEM
• LIMITATION
• PROPOSED SYSTEM
• ADVANTAGE
• HARDWARE AND SOFTWARE REQUIREMENTS
• SYSTEM MODELS
• CONCLUSION
• REFERENCES
 INTRODUCTION

• Cloud computing is a type of Internet based computing that provides

shared computer processing resources and data to computers and other
devices on demand.
•It is a model for enabling ubiquitous, on-demand access to a shared
pool of configurable computing resources (e.g., computer networks,
servers, storage, applications and services).
• Cloud computing relies on sharing of resources to achieve coherence
and economy of scale, similar to a utility (like the electricity grid) over
an electricity network.
 Cloud computing allows companies to avoid up-front infrastructure costs
(e.g., purchasing servers) and it enables organizations to focus on their
core businesses.

 It allows enterprises to get their applications up and running faster, with

improved manageability and less maintenance, and enables
Information technology (IT) teams .

 Among all the benefits scheduling of Virtual Machines is one such issue.
An efficient scheduling algorithm should satisfy user’s QoS constraints
such as response time, cost, energy efficiency etc.
 LITERATURE SURVEY
[1] Prashant Lakkadwala, PriyeshKanungo, et al. [1],analyzed the cloud
Performance based on Evaluation of resource scheduling and compared the
results of Time Shared and Space Shared Techniques for Cloud Resource
Scheduling.

 Cloud provides the resources based on the SaaS model. Cloud systems are developed
based on the concept of shareable resource which can share their computational
resources from the other service providers in the cloud federation.

 The resource scheduling strategies focus on the efficient utilization of resources and
improving the performance of cloud based services.
 The existing work describes the two methods of resource scheduling namely, time
shared and space shared. This scheduling policy, a single resource is sharable among
multiple tasks at the same time.

 The resources switches the jobs after a given time slot. After completing the
processing of a job, the resource is allocated to the next process in the queue.

 The objective of the system is simply to find the improved extendable technique for
resource scheduling. Experimental studies are conducted with both these approaches.
[2] Elhossiny Ibrahim, Nirmeen A. El-Bahnasawy and Fatma A. Omara, et
al. ,describes about the Dynamic Task Scheduling in Cloud Computing Based on
the Availability Level of Resources.

 Virtualization hides the heterogeneity of the resources by introducing its services in the
form of shared pool of configurable resources called Virtual Machines (VMs).It is used
to deliver different IT services to the customers which meet their requirements.

 The service provider hires resources from the resource provider and introduces the
required resources in the form of VM instances which meet the consumers’
requirements.
The following steps to be scheduled on the VMs:

 Resource discovering and filtering ,Resource selection ,Task submission.

There are two types of task scheduling in the Cloud Computing:

 Static scheduling - pre-fetching and pipelining with less runtime .

 Dynamic scheduling- job components/task is unknown beforehand.

 The proposed algorithm is focused on how to dynamically schedule the new

incoming tasks during the execution of other tasks by choosing the best available
resource (i.e., VM).
[3] Sunita Sharma, Nagendra Kumar et al., describes about Task Scheduling
Algorithm in Cloud Computing based on Power Factor.

 A very efficient computing environment is provided by cloud computing where the

customers or several tenants are in need of multiple resources to be provided as a
service over the internet.

 The challenge are how efficiently and effectively the underlying computing resources
like virtual machines, network, storage units, and bandwidth etc. should be managed so
that no computing device is in under-utilization or overutilization state in a dynamic
environment.
 A good task scheduling technique is always required for the dynamic allocation of
the task to avoid such a situation.

 Through this paper we are going to present the new Algorithm based on task
scheduling technique, which will distribute the load effectively among the virtual
machine so that the overall response time (QoS) should be minimal. A comparison of
this proposed Algorithm of task scheduling technique is performed on CloudSim
simulator which shows that, this will outperform the existing techniques like FCFS,
SJF and Genetic Model techniques.
[4] Elhossiny Ibrahim, Nirmeen A El-Bahnasawy and Fatma A Omara, et al. , provides
Job Scheduling based on Harmonization between the Requested and Available
Processing Power in the Cloud Computing Environment.

 The main advantages of the cloud is that servers and IT equipments are collected in a
single place or distributed over different data centers and can be access remotely by
using mobile, PC, laptop, PDA. The main issues are resource management and task
scheduling.

 The main function of the Cloud providers is to provide the services to the users by

deploying VM and defining scheduling policies to users.

 In this paper, a new task scheduling algorithm has been proposed. It is based on
calculating the total processing power of the available resources (i.e., VMs) and the
total requested processing power by the users’ requests, then allocating a group of
the user’s requests to each VM according to the ratio of its needed power
corresponding to the total processing power of all VMs.

 To evaluate the performance of the proposed algorithm, a comparative study has

been done among the proposed algorithm, the existed GA, and PSO algorithms.

 The experimental results by minimizing the total execution time of the user’s
requests and increasing the utilization of the resources.
[5] V.Vinothina, R.Sridaran, PadmavathiGanapathi , provides a Survey on Resource
Allocation Strategies in Cloud Computing.

 Cloud computing has become a new age technology that has got huge potentials in
enterprises and markets and make it possible to access applications and associated data
from anywhere.

 Companies are able to rent resources from cloud for storage and other computational
purposes so that their infrastructure cost can be reduced significantly. Further they can
make use of company-wide access to applications, based on pay-as-you-go model.
 Hence there is no need for getting licenses for individual products. However one of
the major pitfalls in cloud computing is related to optimizing the resources being
allocated.

 Because of the uniqueness of the model, resource allocation is performed with the
objective of minimizing the costs associated with it. The other challenges of resource
allocation are meeting customer demands and application requirements.

 It is believed that this paper would benefit both cloud users and researchers in
overcoming the challenges faced.
 [6] Amer Al-Rahayfeh, Saleh Atiewi, Abdullah Abuhussein and MuderAlmiani ,
Novel Approach to Task Scheduling and Load Balancing Using the Dominant
Sequence Clustering and Mean Shift Clustering Algorithms.

 Cloud computing (CC) is fast-growing and frequently adopted in IT environments due

to the benefits it offers. Task scheduling and load balancing are main under CC.

 To overcome the shortcomings we propose a novel approach that uses dominant

sequence clustering (DSC) for task scheduling and a weighted least connection (WLC)
algorithm for load balancing.
 First, user’s tasks are clustered using the DSC algorithm, After task clustering, each task is
ranked using Modified Heterogeneous Earliest Finish Time (MHEFT) algorithm, where
the highest priority task is scheduled first.

 VM are clustered using a mean shift clustering (MSC) algorithm using kernel functions.
Load balancing is subsequently performed using a WLC algorithm.

 A highly weighted or least connected server is selected for task allocation. Finally, we
evaluate the proposed architecture using metrics such as response time, make span,
resource utilization, and service reliability.
 EXISTING SYSTEM

❑ Load balancing of tasks on virtual machines based on Nature-inspired algorithm has

turned into a field of greater research interest.

 Algorithm for load balancing based on honey bee behavior (LBA_HB) was presented
to decrease the processing time and overall response time. It considers only the number
of tasks handled by VMs for balancing the load.

 This research proposes an Improved Algorithm Based on Honey Bee behavior

(ILBA_HB) by considering the various QoS parameters including service response
time, reliability, cost, and estimate the power consumption of VMs on a host to enhance
load balancing.
 As the situation remains uncertain, fuzzy logic can be applied at various levels of
input to achieve the desired tradeoff between power consumption and QoS.

 The main goal of the proposed study calculates the power consumption of VMs
and considers the important QoS factors to choose the appropriate host and virtual
machine to receive the task.

 The performance of ILBA_HB algorithm is compared with that of LBA_HB in

terms of makespan, response time, degree of imbalance.
LIMITATIONS

 Premature convergence in the later search period.

 The accuracy of the optimal value cannot meet the requirements
sometimes.
PROPOSED SYSTEM

 The explosive growth of cloud computing in present will lead to a massive

increase in both the amount of traffic and the number of service requests to cloud
servers.

 This growth will cause overload on the server. That overload will lead to the
degradation of the performance of servers.

 In order to meet this challenge, efficient load balancing technique is required.

 Load balancing is the process of distributing load on various nodes providing best
resource utilization when nodes are overloaded.
 PROPOSED SYSTEM
 Load balancing includes optimal utilization of resources which increases the
performance of the system and thereby the minimization of the load of competent
resources.

 The objective is to analyze the issues of existing load balancing algorithms and
proposes a Multi objective load balancing transformation technique for improving
the response time of VMs on cloud computing to improve performance for end-
user and also utilizes Cuckoo Search algorithm by considering the various QoS
parameters.

 The Proposed algorithm improves the response times and processing time of the
cloud data center.
ADVANTAGES
 1. The Proposed System provides the efficient resource sharing in the
cloud.
 2. Optimized Load balancing technique is utilized for efficient load
balancing which improves the efficiency with the limited number of
requests queued for delivery
 3. The System Improves processing time and response time of hubs
cloud compared to existing algorithms.
 4. The Cuckoo search Algorithm has strong robustness, high flexibility,
easier to application and fewer tuning parameters
HARDWARE REQUIREMENTS

 This section gives the details and specification of the

hardware on which the system is expected to work.

 Processor : Intel Pentium
 Hard Disk : 60 GB
 Monitor : VGA Color
 RAM : 512 MB
SOFTWARE REQUIREMENTS

 This section gives the details and specification of the software on

which the system is expected to work.
 Front End/GUI Tool : Net Beans 7.0
 Operating System : Windows 07
 Simulation : Cloudsim
 Language : Java
SYSTEM ARCHITECTURE DIAGRAM
DATA FLOW DIAGRAM
ACTIVITY DIAGRAM

User requests

Resource Scheduling

Resource Sharing

Cloud Service Broker

VM Allocation Policy
MODULE DESCRIPTION
 Cloud Provider Management

Cloud Provider management includes both datacenter and user

details management.
 Data Center Configuration

The Data center module simulates an Infrastructure as a Service

(IaaS) provider with an arbitrary number of datacenters. Each
datacenter is entirely customizable.
 Customer Configuration

The customers of the IaaS provider are also simulated and entirely customizable. The user
can set the number of virtual machines each customer owns, a broker responsible for
allocating these virtual machines and resource consumption algorithms.

 Report Generation

It generates reports of each simulation and provides the graphical results on the cloud
performance by the measureable parameters including cost estimation, utilization and
processing performance of datacenters.
 CONCLUSION
• This paper presented a comparative analysis in terms of performance among various load
balancing techniques like Weighted First-come-First-Served (WFCFS), RR, FCFS and
Throttled Load balancer (TLB) and by considering metrics like Energy consumption,
Execution and Response time and Processing.

• Load balancing includes optimal utilization of resources increases the performance of the
system and thereby the minimization of the load of competent resources.

• Multi objective load balancing transformation technique for improving the response time of
VMs on cloud computing to improve performance for end-user and also utilizes Cuckoo
Search algorithm by considering the various QoS parameters including service response time,
reliability, cost, and estimate the power consumption of VMs on a host to enhance load
balancing.
REFERNCES

[1] Amer Al-Rahayfeh, Saleh Atiewi, Abdullah Abuhusseinand

MuderAlmiani, Novel Approach to Task Scheduling and Load
Balancing Using the Dominant Sequence Clustering and
Mean Shift Clustering Algorithms, Future Internet 2019.

[2] Amanpreet Kaur, Birkampal Kaur, “Load balancing

optimization based on hybrid Heuristic-Metaheuristic
techniques in cloud environment”, Journal of King Saud
University - Computer and Information Sciences, March 2019
[3] DaliborKlusácek and Hana Rudová, “Alea 2 – Job Scheduling
Simulator”, SIMUTools 2010 March 15–19, Torremolinos,
Malaga, Spain. Copyright 2010 ICST, ISBN 78-963-9799-87-5

[4] Elhossiny Ibrahim, Nirmeen A. El-Bahnasawy and Fatma A.

Omara, “ Dynamic Task Scheduling in Cloud Computing Based
on the Availability Level of Resources”, International Journal of
Grid and Distributed Computing Vol. 10, No. 8 (2017), pp.21-36