IRJET-Hybridization of algorithms for Cloud Computing

Published on January 2017 | Categories: Documents | Downloads: 63 | Comments: 0 | Views: 202
of 5
Download PDF   Embed   Report

Comments

Content

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395 -0056

Volume: 02 Issue: 05 | Aug-2015

p-ISSN: 2395-0072

www.irjet.net

Hybridization of algorithms for Cloud Computing
Loopy Bhatti1, Gureshpal Singh2, Sanjeev Mahajan3
M.Tech Scholar, Computer Science and Engg., B.C.E.T Gurdaspur, Punjab, India
Associate Professor, Information and Technology, B.C.E.T Gurdaspur, Punjab, India
3 Associate Professor, Computer Science and Engg., B.C.E.T Gurdaspur, Punjab, India
---------------------------------------------------------------------***--------------------------------------------------------------------1. Introduction
Abstract - “Cloud Computing” is a term, which involves
1

2

virtualization, distributed computing, networking, software

Cloud computing is well-known as a provider of vibrant

and web services. A cloud consists of several elements such

services using very large scalable and virtualized

as clients, data center and distributed servers. It consists of

resources above the Internet. Various definitions and

various advantages like fault tolerance, high availability,

interpretations of “clouds” or “cloud computing” exist.

scalability, flexibility, reduced overhead for users by

With fastidious respect to the different usage scopes the

reducing the cost of ownership, on demand services etc.

term is engaged to, we will try to give a agent (as opposed

Cloud computing can be described as a model of Internet-

to complete) set of definitions as proposal towards future

based computing due to Internet based development and

usage in the cloud computing linked research space. We

utilization of computer technology. Scheduling is a critical

try to capture a summary term in a way that best

problem in Cloud computing, because a cloud service

represents the technical aspects and issues related to it. In

provider has to serve many users in Cloud Computing

its broadest form, we can define a 'cloud' is a flexible

System. So job scheduling is the main issue in establishing

execution environment of resources concerning multiple

Cloud Computing Systems. The main goal of scheduling is to

stakeholders and providing a metered service at multiple

maximize the resource utilization, to reduce waiting time,

granularities for a individual level of quality of service. To

execution time. In this thesis, an efficient Hybrid scheduling

be more precise, a cloud is a policy or infrastructure that

approach has been proposed in computational cloud.

enables implementation of code (services, applications

Proposed work is grouping the tasks before resource

etc.), in a managed and elastic fashion, whereas “managed”

allocation according to job priority to reduce the

means that consistency according to pre defined quality

communication overhead. Here tasks are grouped together

parameters is routinely ensured and “elastic” implies that

based on the chosen resources characteristics, to maximize

the resources are put to use according to actual current

resource utilization and minimize processing time. Hence in

requirements

this thesis, we have specifically focused on improving

definitions – implicitly, elasticity includes both up- and

computational cloud performance in terms of CPU

downward scalability of resources and data, but also load-

utilization time, Executed task and Response time. A

balancing of data throughput.

simulation of proposed algorithm is conducted on real time

1.1 Job scheduling

observing

overarching

requirement

cloud server. Experimental results show that proposed
hybrid algorithm performs better than FCFS and Priority

Job scheduling issues are fundamental which identify with

algorithms.

the

Key Words: Scheduling, FCFS, ROUND ROBIN, Priority,

framework. Job scheduling will be a mapping component

Cloud Computing

effectiveness

of

the

entire

cloud

computing

from client's assignments to the proper determination of
assets and its execution. Job scheduling is adaptable and

© 2015, IRJET

ISO 9001:2008 Certified Journal

Page 897

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395 -0056

Volume: 02 Issue: 05 | Aug-2015

p-ISSN: 2395-0072

www.irjet.net

helpful. Jobs and job streams can be planned to run at

enhance the QoS of task scheduling indirectly in a

whatever point needed, taking into account business

cloud environment.

capacities, needs, and needs. Job streams and procedures

v.

The

throughput

of

the

system-Mainly

for

can set up every day, week after week, month to month,

distributed computing frameworks, throughput is a

and yearly ahead of time, and keep running on-interest

measure

jobs without need for help from support staff.

streamlining execution, and it is likewise an objective

1.2 Need of job scheduling

which must be considered in plan of action

The objective of job scheduling in Cloud computing is give
ideal tasks scheduling for clients,

and to give good

throughput and QoS at the same time. Subsequent are the
needs of job scheduling in cloud computing:
i.

has nearly related with one another in the cloud
environment, task planning system capable for the
matching

of

tasks

and

assets. Task

scheduling algorithm can keep up load balancing. So
load balancing get to be another imperative measure
in the cloud.
ii.

computing

and

distributed

storage

administrations, asset interest for clients and assets
supplied by supplier to the clients in such a route
along these lines, to the point that quality of service
can be accomplished. At the point when job
scheduling administration comes to job assignment,
it is important to ensure about QoS of assets.
iii.

Economic Principles-Cloud computing assets are
generally conveyed all through the world. These
assets may fit in with diverse associations. They have
their own particular administration strategies. As a
plan of action, distributed computing as indicated by
the

distinctive

prerequisites,

give

applicable

administrations. So the demand charges are sensible.
iv.

The best running time -jobs can be partitioned into
diverse classes as indicated by the needs of clients,
and after that set the best running time on the
premise of distinctive objectives for every job. It will

© 2015, IRJET

undertaking

planning

advancement. Build throughput for clients and cloud
suppliers would be advantage for both of them.

2. Proposed Work

cloud providers and cloud users. On one hand, providers
hold massive computing resources in their large
datacenters and rent resources out to users on a per-usage
basis. On the other hand, there are users who have
applications with fluctuating loads and lease resources
from providers to run their applications. First, a user
sends a request for resources to a provider. When the
provider receives the request, it looks for resources to

Quality of Service-The cloud is primarily to give
clients

framework

In cloud computing environments, there are two players:

Load Balance-Load balancing and task scheduling

optimal

of

satisfy the request and assigns the resources to the
requesting user. Then the user uses the assigned resources
to run applications and pays for the resources that are
used. When the user’s job is completed then the resources
are free and returned to the service provider. Proper
scheduling is needed to meet user’s requirements and
satisfies the Qos parameters. In this thesis, an efficient
hybrid scheduling approach is proposed in computational
cloud. Proposed work is grouping the tasks before
resource allocation according to job priority to reduce the
communication overhead. The purpose work has been
divided into three sessions namely job creation, system
creation and the schedule.
Following are the metrics on the basis of which results are
evaluated:1. CPU Utilization: - The CPU time is measured in clock
ticks or seconds. Often, it is useful to measure CPU time as
a percentage of the CPU's capacity, which is called the CPU
usage.

ISO 9001:2008 Certified Journal

Page 898

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395 -0056

Volume: 02 Issue: 05 | Aug-2015

p-ISSN: 2395-0072

www.irjet.net

2. Response time:- The elapsed time between the end of
an inquiry or demand on a cloud and the beginning of a
response is called the Response time.
3. Time to complete a batch of jobs:-This is the total
time taken by the cloud to complete the execution of
submitted batch of jobs from beginning to ending by the
different algorithms. To analyze the time to complete the
execution of submitted jobs, first we select 10 jobs from
the selection part and initialize the server configuration.
Fig -1: Proposed algorithm
To analyze the performance of proposed algorithm, 48
simulations have been performed in the cloud and results
are obtained. First of all, the cloud is started by choosing
the configuration and the jobs are created with different
requirements. During the simulation, first we select ten
jobs, then the system checks all the selected jobs that are
Chart -1: CPU Utilization verses number of jobs.

to be executed by the scheduler. It creates the groups of
jobs for execution and set the priority on the basis of CPU
utilization by each group. Now we have task groups to
execute by the scheduler. Here hybrid algorithm performs
their function to optimize the execution of jobs. Then
priority algorithm will sort the groups according to their
priority. The priority is set on the basis of system
threshold value which is evaluated on the basis of CPU
utilization. All the jobs in each group will be executed by
FCFS algorithm. When the jobs under each group are

Chart -2: Response time versus number of jobs

executed completely then performance parameters are
evaluated.
3. CONCLUSIONS
Job scheduling is an essential requirement in cloud
computing environment with the given constraints. Some
intensive researches have been done in the area of job
scheduling of cloud computing. The scheduling algorithms
should order the jobs in a way where balance between
improving the performance and quality of service and at

Chart -3: Total Time to complete verses Jobs

the same time maintaining the efficiency and fairness
among the jobs. This thesis proposed the solution to

© 2015, IRJET

ISO 9001:2008 Certified Journal

Page 899

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395 -0056

Volume: 02 Issue: 05 | Aug-2015

p-ISSN: 2395-0072

www.irjet.net

scheduling problem based on FCFS and priority based

[7] Kushal Dutta ,Ramendu Bikash Guin, Sourav Banerjee,

algorithms. From the experimental results, it has been

Sayan Chakrabarti, Utpal Biswas(2012), “A Smart Job

proved that Proposed Hybrid is more efficient than FCFS,

Scheduling System For Cloud Computing Service

Priority and other hybrid algorithms. Results show that

Providers and Users:Modeling and Simulation” Ist

this algorithm not only improve the Response time but

International

also reduces the total time to complete all the jobs. This

Information Technology(RAIT),pp.346-351, March 2012.

algorithm is more powerful and can be used in dynamic

[8] Ying Chang-tian, Yu Jiong(2012), “Energy-aware

applications.

Genetic Algorithms for Task Scheduling in Cloud

REFERENCES

Computing”,

[1] Omer khalid, Ivo Maljevic, Richard Antony, Miltos

Conference

IEEE

on

Seventh

Recent

Advances

China

Grid

in

Annual

Confrence,pp.43-48,Sep. 2012.

Petridis, Kevin Parott, Markus Schulz (2010), “Deadline

[9] Bo Yang, Xiao fei, Feng Tan, Dong Ho Park(2011),“An

Aware Virtual Machine Scheduler for Grid and Cloud

Utility–Based Job Scheduling Algorithm for Cloud

Computing” IEEE 24'th International conference on

Computing

Advanced Information Networking and Applications

International

Workshop, pp: 85-90 , April 2010.

Computing(CSC), pp.95-102, Dec. 2011.

[2]

Monika

Choudhary,

Sateesh

kumar

considering

Reliability

Conference

on

Factor”IEEE

Cloud

and

Service

peddoju

(2012),“Dynamic Optimization Algorithm for Task
Scheduling in Cloud Environment” International Journal
of Engineering Research and Applications (IJERA)Vol. 2,

[10] Gaochao Xu, Junjie Pang, and Xiaodong Fu,(2013), “ A
Load Balancing Model Based on Cloud Partitioning for
the

Public

Cloud”

IEEE

Transactions

on

Cloud

Computing,Vol.18, No.1,pp.34-39,Feb. 2013.

Issue 3, pp.2564-2568 May-Jun 2012.
[3]Mrs.S.Selvarani,Dr.G.SudhaSadhasivam(2010),“Improv

[11] Pinal Salot (2013),“A Survey of Various scheduling

ed Cost-Based Algorithm for Task Scheduling In Cloud

algorithm

Computing”

International Journal of Research in Engineering and

IEEE

International

Conference

on

Computational Intelligence and Research(ICCIC), pp.1-5,

in

cloud

computing

environment”

Technology (IJRET) Vol.2, Issue.2, pp.131-135,Feb. 2013.

Dec 2010.
[4] Jaspreet kaur (2012), “Comparison of load balancing

[12] T.Desai, J. Prajapati(2013),“A Survey Of Various

algorithms

Load Balancing Techniques And Challenges In Cloud

in

a Cloud”

International

Journal

of

Engineering Research and Applications (IJERA) Vol. 2,

Computing”

Issue 3, pp.1169-1173, May-Jun 2012.

Technology Research, VoL.2, Issue.11, pp.158-161, Nov.

[5] Supreeth S, Shobha Biradar, (2013), “Scheduling

2013.

Virtual Machines for Load Balancing in Cloud
Computing Platform” International Journal of Science
and Research (IJSR), Vol.2,Issue:6, pp: 437-441,june 2013.
[6] Jinho Hwang, Timothy Wood (2012), “Adaptive
Dynamic Priority Scheduling for Virtual Desktop
Infrastructures” in the proceedings of

International

Journal

of

Scientific

&

[13] Luqun Li, “An Optimistic Differentiated Service Job
Scheduling System for Cloud Computing Service Users
and Providers”, IEEE Third International Conference on
Multimedia and Ubiquitous Engineering, pp.295-299,4-6
june 2009.

IEEE 20th

International Workshop on Quality of Services,4-5 June

[14]

2012.

Scheduler for the Cloud Computing Systems”, IEEE

© 2015, IRJET

Yujia

ISO 9001:2008 Certified Journal

Ge,

Guiyi

Wei(2010),“GA-Based

Task

Page 900

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395 -0056

Volume: 02 Issue: 05 | Aug-2015

p-ISSN: 2395-0072

www.irjet.net

International Conference on Web Information Systems
and Mining,pp.181-186,24-25 Oct 2010.
[15] Amandeep Kaur Sidhu, Supriya Kinger(2013), “A
Sophisticated Approach for Job Scheduling in Cloud
Server”International Journal of Computer Trends and
Technology (IJCTT) – volume 4 Issue 7–July 2013.

© 2015, IRJET

ISO 9001:2008 Certified Journal

Page 901

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close