Joint Pricing and Capacity Planning in the IaaS

Cloud Market
Abstract
In the cloud context, pricing and capacity planning are two important factors
to the profit of the Infrastructure-as-a-Service(IaaS) providers. This paper
investigates the problem of joint pricing and capacity planning in the IaaS provider
market with a setof Software-as-a-Service (SaaS) providers, where each SaaS
provider leases the virtual machines (VMs) from the IaaS providers toprovide
cloud-based application services to its end-users. We study two market models,
one with a monopoly IaaS provider market, theother with multiple-IaaS-provider
market. For the monopoly IaaS provider market, we first study the SaaS providers’
optimal decisionsin terms of the amount of end-user requests to admit and the
number of VMs to lease, given the resource price charged by the IaaSprovider.
Based on the best responses of the SaaS providers, we then derive the optimal
solution to the problem of joint pricing andcapacity planning to maximize the IaaS
provider’s profit. Next, for the market with multiple IaaS providers, we formulate
the pricing andcapacity planning competition among the IaaS providers as a three-
stage Stackelberg game. We explore the existence and uniquenessof Nash
equilibrium, and derive the conditions under which there exists a unique Nash
equilibrium. Finally, we develop an iterativealgorithm to achieve the Nash
equilibrium.
Architecture Diagram

Existing System
Cloud computing provides an attractive paradigm forthe dynamic
provisioning of computing services in a"pay-as-you-go" manner. These services
are typicallyclassified into three categories: Infrastructure as-a-Service, Platform-
as-a-Service andSoftware-as-a-Service .With IaaS, such as AmazonEC2, each
physical machine is virtualized into multiplevirtual machines, and the computing
resourcesare leased to cloud costumers in the form of VMs. WithPaaS, such as
Google App Engine, a computing platformis delivered on which cloud users can
develop and runtheir applications. With SaaS, the applications can beaccessed over
the Internet by end-users without softwarerelated cost and effort.

Proposed System
Cloud computing provides an attractive paradigm forthe dynamic
provisioning of computing services in a"pay-as-you-go" manner. These services
are typicallyclassifiedinto three categories : Infrastructureas-a-Service (IaaS),
Platform-as-a-Service (PaaS), andSoftware-as-a-Service (SaaS).With IaaS, such as
AmazonEC2, each physical machine is virtualized into multiplevirtual machines
(VMs), and the computing resourcesare leased to cloud costumers in the form of
VMs. WithPaaS, such as Google App Engine, a computing platformis delivered on
which cloud users can develop and runtheir applications. With SaaS, the
applications can beaccessed over the Internet by end-users without software
Related cost and effort. In general, SaaS providers utilizethe internal resources of
their own data centers or rentresources from a specific IaaS provider.The end-users
can get access to the applicationsprovided by SaaS providers over the Internet.
Toserve their customers, SaaS providers lease computingresources from IaaS
providers. The scenario we considerin this paper is that SaaS providers can not buy
toestablish their own data centers to server users’ requests and thus lease resources
from IaaS providers withoutupfront investment in infrastructure and software.

Module Implementation
 Optimal Decisions Of The SaaS Providers
 Joint Pricing And Capacity Planning
 Analysis Of The Three-Stage Stackelberg Game
 Extension To The Market With Heterogeneous IaaS Providers
 Analysis of the Numerical Results

Optimal Decisions Of The SaaS Providers

 In this section, we derive the optimal decisions of theSaaS providers for a
given VM price in the monopolyIaaS provider market.

Joint Pricing And Capacity Planning

  Based on the optimal decisions of the SaaS providers,in this section we
study the joint pricing and capacityplanning for the IaaS provider
tomaximize its profitin the monopoly IaaS provider market.
 We assumethat the SaaS providers send the information ofguaranteed
maximum average response time dis and theutility levels to the IaaS
provider.
 They competewith each other in terms of prices and availablecapacities in
order to maximize their profit. We assumethat both the IaaS and SaaS
providers are selfish andrational.
 Observe that the interaction between the IaaSand SaaS providers is a typical
leader-follower gamethat can be analyzed by using the Stackelberg
gameframework.
 Specifically, we cast this interaction as athree-stage Stackelberg game,
where the IaaS and SaaSproviders adapt their decisions dynamically to reach
anequilibrium point.
Analysis Of The Three-Stage Stackelberg Game
 We analyze the subgame perfect equilibrium of the threestageStackelberg
game by using the backward inductionmethod.
 Firstly, we study the SaaS providers’ optimalresponses at Stage Three, given
the IaaS providers’ pricesand available capacities. Secondly, we focus on
StageTwo and analyze the IaaS providers’ optimal decisions
Extension To The Market With Heterogeneous IaaS Providers
 We assumed that IaaS providers arehomogeneous, and the maximum
available capacity issame for each IaaS provider. In this section, weextend to
the case with heterogeneous IaaS providers,where the maximum number of
servers in the IaaSproviders are denoted by X1,X2, . . .,XN, respectively.
Analysis of the Numerical Results
  In this section, we first compare the profit achieved byour proposed scheme
to the scheme with fixed capacityin the monopoly IaaS provider market.
Then we discussthe Nash equilibrium in the market with multiple
IaaSproviders, and evaluate the performance achieved at theequilibrium.

Algorithm
 Iterative Algorithm
 We present an iterative algorithm for each IaaSprovider to reach the
Nash equilibrium of the dynamic.

System Requirements

H/W System Configuration:-

Processor - Pentium –III

Speed - 1.1 Ghz

RAM - 256 MB(min)

Hard Disk - 20 GB

Key Board - Standard Windows Keyboard

Mouse - Two or Three Button Mouse

Monitor - SVGA

S/W System Configuration

 Operating System :Windows95/98/2000/XP /7

 Application Server : Tomcat5.0/6.X /8.X

  Front End : HTML, Java, Jsp

 Scripts : JavaScript,jquery,ajax

 Server side Script : Java Server Pages.

 Database Connectivity : Mysql.

Conclusion
This paper studied joint pricing and capacity planningin the IaaS provider
market, where each SaaS providerleases VMs from the IaaS providers to provide
cloudbasedapplication services to its end-users. For themonopoly IaaS provider
market, we first formulated theprofit maximization problem for each SaaS
provider, andderived its optimal decisions in terms of the amount ofend-user
requests to admit and the number of VMs tolease. Based on the best responses of
the SaaS providers,we then derived the optimal solution to the problem ofjoint
pricing and capacity planning to maximize the IaaSprovider’s profit. Next, for the
market with multiple IaaSproviders, we formulated the pricing and capacity
planningcompetition among the IaaS providers as a threestageStackelberg game.
We characterized the conditionsunder which there exists a unique Nash
equilibrium,and developed an iterative algorithm to achieve theequilibrium.For
future work, it would be interesting to consider theheterogenous case where the
VM instances provided byeach IaaS provider are different. Also, it would be
interestingto consider the case where the VM migration andconsolidation are
exploited to increase IaaS providers’profit further.