ISSN 2348–2370

Vol.08,Issue.17,
October-2016,
Pages:3367-3372

www.ijatir.org

A Profit Maximization Scheme with Guaranteed Quality of Service

in Cloud Computing
BADITHALA VANITHA1, SABAVATH.RAJESHWAR2, DR. MANDALAPU.EKAMBARAM NAIDU3
1
PG Scholar, Dept of CSE(SE), Arjun College of Technology and Sciences, Hyderabad, TS, India.
2
Associate Professor & HOD, Dept of CSE, Arjun College of Technology and Sciences, Hyderabad, TS, India.
3
Professor, Dept of CSE, Arjun College of Technology and Sciences, Hyderabad, TS, India.

Abstract: A fruitful and successful way to deal with give
enlisting resources and organizations to customers on
interest, appropriated registering has ended up being
progressively popular. From cloud organization suppliers'
perspective, advantage is a champion amongst the most basic
thoughts, and it is generally controlled by the outline of a
cloud organization stage under given business segment
demand. In any case, a lone whole deal renting arrangement
is regularly gotten to orchestrate a cloud stage, which can't
guarantee the organization quality yet prompts honest to
goodness resource waste. In this paper, a twofold resource
renting arrangement is sketched out firstly in which transient
renting and whole deal renting are joined going for the
present issues. This twofold renting arrangement can
sufficiently guarantee the way of organization of all sales
and reduction the advantage misuse amazingly. Also, an
organization structure is considered as a M/M/m+D lining
model and the execution markers that impact the advantage
of our twofold renting arrangement are inspected, e.g., the
ordinary charge, the extent of sales that need break servers,
and so forth. Thirdly, an advantage expansion issue is
characterized for the twofold renting arrangement and the
updated configuration of a cloud stage is gotten by handling
the advantage enhancement issue. Finally, a movement of
calculations is coordinated to take a gander at the event of
our proposed arrangement with that of the single renting
arrangement. The results show that our arrangement can't
simply guarantee the organization way of all requesting,
moreover get more advantage than the last said.
Keywords: Cloud Computing, Guaranteed Service Quality,
Multi-Server System, Profit Maximization, Queuing Model,
Service-Level Agreement, Waiting Time.
I. INTRODUCTION
Conveyed registering is quickly transforming into a
practical and beneficial technique for figuring resources. By
joined organization of advantages and organizations,
dispersed registering passes on encouraged organizations
over the Internet conveyed registering can give the most
fiscally sharp and imperativeness compelling technique for
figuring resources administration. Distributed processing
change's information advancement into ordinary items and
utilities by using the compensation per-use assessing model.
An organization supplier rents resources from the base
dealers, produces reasonable multi server structures, and
gives distinctive organizations to customers. A customer
exhibits an organization requesting to an organization
supplier, the fancied result from the organization supplier
with certain organization level understands. At that point
pays for the organization in light of the measure of the
organization and the way of the organization. An
organization supplier can make differing multi server
structures for different application spaces, such that
organization sales of different natures are sent to different
multi server systems. Owing to redundancy of PC structure
frameworks and limit system cloud may not be trustworthy
for data, the security score is concerned. In disseminated
figuring security is greatly upgraded as a consequence of a
prevalent advancement security system, which is presently
successfully open and direct. Applications no more continue
running on the desktop Personal Computer however continue
running in the cloud.
This infers the PC does not require the taking care of force
or hard hover space as asked for by standard desktop
programming. Viable servers et cetera are not any more
required. The handling power of the cloud can be used to
supplant or supplement internal figuring resources.
Affiliations no more need to purchase enrolling advantages
for handle the utmost tops. Distributed computing is rapidly
turning into a compelling and productive method for
registering assets. By brought together administration of
assets and administrations, distributed computing conveys
facilitated administrations over the Internet. Distributed
computing can give the most financially savvy and vitality
productive method for figuring assets administration.
Distributed computing transform's data innovation into
normal things and utilities by utilizing the pay-per-use
estimating model an administration supplier rents assets
from the foundation sellers, fabricates suitable multi server
frameworks, and gives different administrations to clients. A
customer presents an administration solicitation to an
administration supplier, gets the wanted result from the
administration supplier with certain administration level
assertion.

Copyright @ 2016 IJATIR. All rights reserved.

 BADITHALA VANITHA, SABAVATH.RAJESHWAR, DR. MANDALAPU.EKAMBARAM NAIDU
At that point pays for the administration taking into
account the measure of the administration and the nature of
the administration. An administration supplier can fabricate
diverse multi server frameworks for various application
areas, such that administration solicitations of various
natures are sent to various multi server frameworks.
Inferable from repetition of PC framework systems and
capacity framework cloud may not be solid for information,
the security score is concerned. In distributed computing
security is massively enhanced on account of a predominant
innovation security framework, which is currently
effortlessly accessible and reasonable. Applications no more
keep running on the desktop Personal Computer yet keep
running in the cloud. This implies the PC does not require
the preparing power or hard circle space as requested by
customary desktop programming. Effective servers and so
forth are no more required. The registering force of the cloud
can be utilized to supplant or supplement inner figuring
assets. Associations no more need to buy registering assets
to handle the limit crests.
II.EXISTING AND PROPOSED SYSTEMS
A. Existing System
In general, a service provider rents a certain number of
servers from the infrastructure providers and builds different
multi-server systems for different application domains. Each
multi-server system is to execute a special type of service
requests and applications. Hence, the renting cost is
proportional to the number of servers in a multi-server
system. The power consumption of a multi-server system is
linearly proportional to the number of servers and the server
utilization, and to the square of execution speed. The
revenue of a service provider is related to the amount of
service and the quality of service. To summarize, the profit
of a service provider is mainly determined by the
configuration of its service platform. To configure a cloud
service platform, a service provider usually adopts a single
renting scheme. That’s to say, the servers in the service
system are all long-term rented. Because of the limited
number of servers, some of the incoming service requests
cannot be processed immediately. So they are first inserted
into a queue until they can handle by any available server.
Disadvantages of Existing System:
 The waiting time of the service requests is too long.
 Sharp increase of the renting cost or the electricity cost.
Such increased cost may counterweight the gain from
penalty reduction. In conclusion, the single renting
scheme is not a good scheme for service providers.
B. Proposed System
In this paper, we propose a novel renting scheme for
service providers, which not only can satisfy quality-of-
service requirements, but also can obtain more profit. A
novel double renting scheme is proposed for service
providers. It combines long-term renting with short-term
renting, which can not only satisfy quality-of-service
requirements under the varying system workload, but also
reduce the resource waste greatly. A multi-server system
International Journal of Advanced Technology and Innovative Research
Volume. 08, IssueNo.17, October-2016, Pages: 3267-3272
adopted in our paper is modeled as an M/M/m+D queuing
model and the performance indicators are analyzed such as
the average service charge, the ratio of requests that need
short term servers, and so forth. The optimal configuration
problem of service providers for profit maximization is
formulated and two kinds of optimal solutions, i.e., the ideal
solutions and the actual solutions, are obtained respectively.
A series of comparisons are given to verify the performance
of our scheme. The results show that the proposed Double-
Quality-Guaranteed (DQG) renting scheme can achieve
more profit than the compared Single-Quality-Unguaranteed
(SQU) renting scheme in the premise of guaranteeing the
service quality completely.
Advantages of Proposed System:
 Since the requests with waiting time D are all assigned
to temporary servers, it is apparent that all service
requests can guarantee their deadline and are charged
based on the workload according to the SLA. Hence, the
revenue of the service provider increases.
 Increase in the quality of service requests and maximize
the profit of service providers.
 This scheme combines short-term renting with long-
term renting, which can reduce the resource waste
greatly and adapt to the dynamical demand of
computing capaczty.
Fig.1. System Architecture.
III. IMPLEMENTATION OF MODULES
1. Cloud computing,
2. Queuing model.
3. Business Service Module
4. Cloud customer Module.
5. Infrastructure Service Provider Module.
1. Cloud Computing
Cloud computing describes a type of outsourcing of
computer services, similar to the way in which the supply of
electricity is outsourced. Users can simply use it. They do
not need to worry where the electricity is from, how it is
made, or transported. Every month, they pay for what they
consumed. The idea behind cloud computing is similar: The
user can simply use storage, computing power, or specially
crafted development environments, without having to worry
how these work internally. Cloud computing is usually
Internet-based computing. The cloud is a metaphor for the
Internet based on how the internet is described in computer
network diagrams; which means it is an abstraction hiding
the complex infrastructure of the internet. It is a style of
 A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Computing
computing in which IT-related capabilities are provided “as a
service”, allowing users to access technology-enabled
services from the Internet ("in the cloud") without
knowledge of, or control over the technologies behind these
servers.
2. Queuing model
We consider the cloud service platform as a multi-server
system with a service request queue. The clouds provide
resources for jobs in the form of virtual machine (VM). In
addition, the users submit their jobs to the cloud in which a
job queuing system such as SGE, PBS, or Condor is used.
All jobs are scheduled by the job scheduler and assigned to
different VMs in a centralized way. Hence, we can consider
it as a service request queue. For example, Condor is a
specialized workload management system for compute
intensive jobs and it provides a job queuing mechanism,
scheduling policy, priority scheme, resource monitoring, and
resource management. Users submit their jobs to Condor,
and Condor places them into a queue, chooses when and
where to run they based upon a policy. An
M/M/m+Dqueuing model is build for our multi-server
system with varying system size. And then, an optimal
configuration problem of profit maximization is formulated
in which many factors are taken into considerations, such as
the market demand, the workload of requests, the server-
level agreement, the rental cost of servers, the cost of energy renting scheme, and the five curves reach the peak at
consumption, and so forth. The optimal solutions are solved
for two different situations, which are the ideal optimal
solutions and the actual optimal solutions.
3. Business Service Providers Module
Service providers pay infrastructure providers for renting
their physical resources, and charge customers for processing
their service requests, which generates cost and revenue,
respectively. The profit is generated from the gap between
the revenue and the cost. In this module the service providers the above five scenarios under the given server speed.
considered as cloud brokers because they can play an
important role in between cloud customers and infrastructure
providers, and he can establish an indirect connection
between cloud customer and infrastructure providers.
4. Infrastructure Service Provider Module
In the three-tier structure, an infrastructure provider the
basic hardware and software facilities. A service provider
rents resources from infrastructure providers and prepares a
set of services in the form of virtual machine (VM).
Infrastructure providers provide two kinds of resource
renting schemes, e.g., long-term renting and short-term
renting. In general, the rental price of long-term renting is
much cheaper than that of short-term renting.
5. Cloud Customers
A customer submits a service request to a service provider
which delivers services on demand. The customer receives
the desired result from the service provider with certain
service-level agreement, and pays for the service based on
the amount of the service and the service quality.
International Journal of Advanced Technology and Innovative Research
Volume. 08, IssueNo.17, October-2016, Pages: 3367-3372
IV. PERFORMANCE COMPARISON
Using our resource renting scheme, temporary servers are
rented for all requests whose waiting time are equal to the
deadline, which can guarantee that all requests are served
with high service quality. Hence, our scheme is superior to
the traditional resource renting scheme in terms of the
service quality. Next, we conduct a series of calculations to
compare the profit of our renting scheme and the renting
scheme in [3]. In order to distinguish the proposed scheme
and the compared scheme, the proposed scheme is renamed
as Double-Quality-Guaranteed (DQG) renting scheme and
the compared scheme is renamed as Single-Quality-
Unguaranteed (SQU) renting scheme in this paper.
A. Profit Comparison under Different Quality-
Guaranteed Ratio
Let λ be 5.99 and the other parameters be the same as
those in Section 5. In the first example, for a given number
of servers, we compare the profit using the SQU renting
scheme with quality-guaranteed ratio 100%, 99%, 92%, 85%
and the optimal profit using our DQG renting scheme.
Because the quality-guaranteed ratios 100% cannot be
achieved using the SQU renting scheme, hence, we set
99.999999%≈100%. The results are shown in Fig. 2. From
the figure, we can see that the profit obtained using the
proposed scheme is always greater than that using the SQU
different sizes. In addition, the profit obtained by a service
provider increases when the quality-guaranteed ratio
increases from 85% to 99%, but decreases when the ratio is
greater than 99%. That is because more service requests are
charged with the increasing ratio from 85% to 99%; but once
the ratio is greater than 99%, the cost to expand the server
size is greater than the revenue obtained from the extra
quality-guaranteed requests; hence, the total profit is
reduced. In the second example, we compare the profit of
TABLE 1. Comparison of the two methods for finding
the optimal size
The results are given in Fig3. The figure shows the trend
of profit when the server speed is increasing from 0.1 to 2.9.
From the figure, we can see that the curves increase firstly
and reach the peak at certain speed, and then decrease along
with the increasing speed on the whole. The figure verifies
that our proposed scheme can obtain more profit than the
SQU renting scheme. Noticed that the changing trends of the
curves of the SQU renting scheme with 100%, 99%, 92%,
and 85% quality-guaranteed ratio are interesting they show
an increasing trend at the beginning and then decrease during
a small range of speed repeatedly. The reason is analyzed as
follows. When the server speed is changing within a small
speed range, in order to satisfy the required deadline-
 BADITHALA VANITHA, SABAVATH.RAJESHWAR, DR. MANDALAPU.EKAMBARAM NAIDU
guaranteed ratio, the number of servers rented by a service
provider keeps unchanged. At the beginning, the added
revenue is more than the added cost, so the profit is
increasing. However, when the speed becomes greater, the
energy consumption increases, leading to the total increased
cost surpassing the increased revenue, hence, the profit
decreases. In the third example, we explore the changing
trend of the profit with different D, and the results are shown
as Fig.4.

TABLE 2. Comparison of the two methods for finding

the optimal speed

Fig.3. Profit versus s and different quality-guaranteed

ratios.
TABLE 3. Comparison of the two methods for finding
Fig.4 (a) gives the numerical results when the server speed
the optimal size and the optimal speed
is fixed at 0.7 and Fig. 4(b) shows the numerical results
when the number of servers is fixed at 5. We analyze the
results as follows From Fig. 4(a), we can see that the profit
obtained using the SQU renting scheme increases slightly
with the increment of D. That is because the service charge
keeps constant but the extra cost is reduced when D is
greater. As a consequence, the profit increases. The second
phenomenon from the figure is that the curves of SQU 92%
and SQU 85% have sharp drop at some points and then
ascend gradually and smoothly.

Fig.2. Profit versus m and different quality-guaranteed Fig.4. Profit versus D and different quality-guaranteed
ratios. ratios.
International Journal of Advanced Technology and Innovative Research
Volume. 08, IssueNo.17, October-2016, Pages: 3267-3272
 A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Computing
Fig.5. Comparison between our schemes with that in [3]
The reasons are explained as follows. When the server
speed is fixed, enough servers are needed to satisfy the given
quality-guaranteed ratio. By calculating, we know that the
number of required servers is the same for all D values in a
certain interval. For example, [6, 8] and [9] are two intervals
of D for the curve of SQU 92%, and the required servers are
10 and 9, respectively. For all D within the same interval,
their costs are the same with each other. Whereas, their
actual quality-guaranteed ratios are different which get
greater with the increasing D. Hence, during the same
interval, the revenue gets greater as well as the profit.
However, if the deadline increases and enters a different
interval, the quality-guaranteed ratio sharply drops due to the
reduced servers, and the lost revenue surpasses the reduced
cost, hence, the profit sharply drops as well. Moreover, we
can also see that the profit of SQU 100% is much less than
the other scenarios. That is because when the quality-
guaranteed ratio is great enough, adding small revenue leads
to a much high cost. From Fig. 4(b), we can see that the
curves of SQU 92% and SQU 85% descend and ascend
repeatedly. The reasons are same as that of Fig. 4(a). The
deadlines within the same interval share the same minimal
speed, hence, the cost keeps constant. At the same time, the
revenue increases due to the increasing quality-guaranteed
ratio. As a consequence, the profit increases. At each break
point, the minimal speed satisfying the required quality-
guaranteed ratio gets smaller, which leads to a sharp drop of
the actual quality-guaranteed ratio hence, the revenue as well
as the profit drops?
B. Comparison of Optimal Profit
In order to further verify the superiority of our proposed
scheme in terms of profit, we conduct the following
comparison between the optimal profit achieved by our DQG
renting scheme and that of the SQU renting scheme in [3]. In
this group of comparisons, λ is set as 6.99,D is 5, r is varying
from 0.75 to 2.00 in step of 0.25, and the other parameters
are the same as Section 5. In Fig. 5, the optimal profit and
the corresponding configuration of two renting schemes are
presented. From Fig. 5(a) we can see that the optimal profit
obtained using our scheme is always greater than that using
the SQU renting scheme. According to the calculation, our
scheme can obtain 4.17 percent more profit on the average
International Journal of Advanced Technology and Innovative Research
Volume. 08, IssueNo.17, October-2016, Pages: 3367-3372
than the SQU renting scheme. This shows that our scheme
outperforms the SQU renting scheme in terms of both of
quality of service and profit. Figs. 5(b) and 5(c) compare the
server size and speed of the two schemes. The figures show
that using our renting scheme the capacity provided by the
long-term rented servers is much less than the capacity using
the SQU renting scheme. That is because a lot of requests are
assigned to the temporary servers using our scheme, and less
servers and slower server speed are configured to reduce the
waste of resources in idle period. In conclusion, our scheme
can not only guarantee the service quality of all requests, but
also achieve more profit than the compared one.
V. CONCLUSION
Maximize the profit of service providers; this paper has
proposed a novel Double-Quality-Guaranteed (DQG) renting
scheme for service providers. This scheme combines short-
term renting with long-term renting, which can reduce the
resource waste greatly and adapt to the dynamical demand of
computing capacity. An M/M/m+Dqueuing model is build
for our multi-server system with varying system size. And
then, an optimal configuration problem of profit
maximization is formulated in which many factors are taken
into considerations, such as the market demand, the
workload of requests, the server-level agreement, the rental
cost of servers, the cost of energy consumption, and so forth.
The optimal solutions are solved for two different situations,
which are the ideal optimal solutions and the actual optimal
solutions. In addition, a series of calculations are conducted
to compare the profit obtained by the DQG renting scheme
with the Single-Quality-Unguaranteed (SQU) renting
scheme. The results show that our scheme outperforms the
SQU scheme in terms of both of service quality and profit.
VI. REFERENCES
[1] Jing Mei, Kenli Li, Member, IEEE, Aijia Ouyang and
Keqin Li, Fellow, IEEE, “A Profit Maximization Scheme
with Guaranteed Quality of Service in Cloud Computing”,
Transactions on Computers, Vol., No., 2015.
[2] K. Hwang, J. Dongarra, and G. C. Fox, Distributed and
Cloud Computing. Elsevier/Morgan Kaufmann, 2012.
[3] J. Cao, K. Hwang, K. Li, and A. Y. Zomaya, “Optimal
multi-server configuration for profit maximization in cloud
 BADITHALA VANITHA, SABAVATH.RAJESHWAR, DR. MANDALAPU.EKAMBARAM NAIDU
computing,” IEEE Trans. Parallel Distrib. Syst., vol. 24, no. Author’s Profile:
6, pp. 1087–1096, 2013. Badithala Vanitha, PG Scholar, Dept of
[4] A. Fox, R. Griffith, A. Joseph, R. Katz, A. Konwinski, G. CSE(SE), Arjun College of Technology
Lee, D. Patterson, A. Rabkin, and I. Stoica, “Above the and Sciences, Hyderabad, TS, India.
clouds: A berkeley view of cloud computing,” Dept.
Electrical Eng. and Comput. Sciences, vol. 28, 2009.
[5] R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg, and I.
Brandic, “Cloud computing and emerging it platforms:
Vision, hype, and reality for delivering computing as the 5th
utility,” Future Gener. Comp. Sy., vol. 25, no. 6, pp. 599– Mr. Sabavath Rajeshwar, received the
616, 2009. Master of Technology degree in
[6] P. Mell and T. Grance, “The NIST definition of cloud Computer Science And Engineering from
computing. national institute of standards and technology,” the Acharya Nagarjuna University, he
Information Technology Laboratory, vol. 15, p. 2009, 2009. received the Bachelor Of Engineering
[7] J. Chen, C. Wang, B. B. Zhou, L. Sun, Y. C. Lee, and A. degree from Swamiramanand Thiede
Y. Zomaya, “Tradeoffs between profit and customer Institute of Science And Technology. He
satisfaction for service provisioning in the cloud,” in Proc. is currently working as Associate Professor and a Head of
20th Int’l Symp. High Performance Distributed Computing. the Department of CSE with Arjun College of Technology
ACM, 2011, pp. 229–238. And Sciences. His interest subjects are Software
[8] J. Mei, K. Li, J. Hu, S. Yin, and E. H.-M. Sha, “Energy Engineering, Database Management System, Software
aware preemptive scheduling algorithm for sporadic tasks on Testing Methodologis,etc. Email: arjuncse2014@gmail.com.
dvs platform,” MICROPROCESS MICROSY., vol. 37, no.
1, pp. 99–112, 2013. Prof Shri. Dr. Mandalapu Ekambaram
[9] P. de Langen and B. Juurlink, “Leakage-aware Naidu, receivedthe Master of Technology
multiprocessor scheduling,” J. Signal Process. Sys., vol. 57, degree in Computer Science Engineering,
no. 1, pp. 73–88, 2009. from the Mangalore University, he
[10] G. P. Cachon and P. Feldman, “Dynamic versus static received the Bachelor of Engineering
pricing in the presence of strategic consumers,” Tech. Rep., degree from S.V University, Tirupathi.
2010. received the Ph.D in Computer Science
[11] Y. C. Lee, C. Wang, A. Y. Zomaya, and B. B. Zhou, Engineering from the University of Mysore. He is currently
“Profitdriven scheduling for cloud services with data access working as Professor of CSE with Arjun College of
awareness,” J. Parallel Distr. Com., vol. 72, no. 4, pp. 591– Technology and Sciences. His interest subjects are Software
602, 2012. Engineering, Database Management System, Software
[12] M. Ghamkhari and H. Mohsenian-Rad, “Energy and Testing Mehodologis, etc, Email:menaidu2005@yahoo.com.
performance management of green data centers: a profit
maximization approach,” IEEE Trans. Smart Grid, vol. 4, no.
2, pp. 1017–1025, 2013.

International Journal of Advanced Technology and Innovative Research

Volume. 08, IssueNo.17, October-2016, Pages: 3267-3272