The document discusses a scalable, stochastic modeling approach for analyzing the availability of large-scale Infrastructure-as-a-Service (IaaS) clouds, addressing common component failures that can violate service level agreements (SLAs). It introduces an interacting Markov chain-based method to reduce analysis complexity, using fixed-point iteration to resolve dependencies among sub-models. The proposed model is shown to effectively handle large IaaS clouds, providing cloud service providers with valuable insights for system design and operational tuning to meet availability SLAs.

1. GLOBALSOFT TECHNOLOGIES
IEEE PROJECTS & SOFTWARE DEVELOPMENTS
IEEE FINAL YEAR PROJECTS|IEEE ENGINEERING PROJECTS|IEEE STUDENTS PROJECTS|IEEE
BULK PROJECTS|BE/BTECH/ME/MTECH/MS/MCA PROJECTS|CSE/IT/ECE/EEE PROJECTS
CELL: +91 98495 39085, +91 99662 35788, +91 98495 57908, +91 97014 40401
Visit: www.finalyearprojects.org Mail to:ieeefinalsemprojects@gmail.com
Scalable Analytics for IaaS Cloud Availability
Abstract—
In a large Infrastructure-as-a-Service (IaaS) cloud, component failures are quite
common. Such failures may lead tooccasional system downtime and eventual
violation of Service Level Agreements (SLAs) on the cloud service availability.
Theavailability analysis of the underlying infrastructure is useful to the service
provider to design a system capable of providing a definedSLA, as well as to
evaluate the capabilities of an existing one. This paper presents a scalable,
stochastic model-driven approach toquantify the availability of a large-scale
IaaS cloud, where failures are typically dealt with through migration of physical
machinesamong three pools: hot (running), warm (turned on, but not ready), and
cold (turned off). Since monolithic models do not scale for largesystems, we use
an interacting Markov chain based approach to demonstrate the eduction in the
complexity of analysis and thesolution time. The three pools are modeled by
interacting sub-models. Dependencies among them are resolved using fixed-pointiteration,
for which existence of a solution is proved. The analytic-numeric
solutions obtained from the proposed approach and from the

2. monolithic model are compared. We show that the errors introduced by
interacting sub-models are insignificant and that our approachcan handle very
large size IaaS clouds. The simulative solution is also considered for the
proposed model, and solution time of themethods are compared.
EXISTING SCHEME
The availability analysis of the underlying infrastructure is useful to the service
provider to design a system capable of providing a definedSLA, as well as to
evaluate the capabilities of an existing one. The three pools are modeled by
interacting sub-models. Dependencies among them are resolved using fixed-point
iteration, for which existence of a solution is proved..Many of the existing
published models are hierarchical in nature. In our case, complexity and
characteristics of large IaaS clouds (e.g., migration of PMs from one
pool to another) lead to cyclic dependency among the submodels,needing fixed-point
iteration.
PROPOSED SCHEME
The analytic-numeric solutions obtained from the proposed approach and from
themonolithic model are compared. We show that the errors introduced by
interacting sub-models are insignificant and that our approachcan handle very
large size IaaS clouds. The simulative solution is also considered for the
proposed model, and solution time of themethods are compared. We state the
availability assessment problem for anIaaScloud and propose a realistic

3. monolithicmodel representative of the state-of-the-arts. an interacting sub-models
approach tosolve the largeness problem of the monolithic availability
model . The overall model solutionis obtained by fixed-point iteration over
individualsub-model solutions. To solve such problems, the hierarchical
composition is introduced in (and many other papersand books), where a two-level
hierarchical model is proposed.Each subsystem is modeled by a Markov
chain andthe system. The specific case of cloud computing, some modeling
approaches focusing on dependability aspects have beenproposed in recent
years. The scalable stochasticapproach that we describe can be complementary
to thiswork as the measured failure/repair rates of hardware componentscan be
used to parameterize the model we propose. an anomaly prediction system
(ALERT) forachieving robust hosting infrastructures is proposed. In special
cases, closed form solutions canalso be derived to solve very large cloud models
quickly.Cloud service providers can benefit from the proposedmodeling
approach during design, development, testingand operation of IaaS cloud.
CONCLUSIONS
This paper describes a stochastic modeling approach foravailability analysis of
large IaaS cloud systems. We showhow scalability issues for a monolithic
model can beresolved by means of interacting sub-models or by means of
simulation. The interacting sub-models approach quicklyprovides model
solutions facilitating scalability without significantlycompromising the
accuracy. Simulation providesresults that closely match with monolithic and
interactingsub-models approaches and, for large systems, results areobtained
faster. In special cases, closed form solutions canalso be derived to solve very
large cloud models quickly.Cloud service providers can benefit from
theproposedmodeling approach during design, development, testingand
operation of IaaS cloud. During design and development,providers can use these

4. models to determine the poolsize required to offer a specific availability SLA.
In the testingand operational stages, the providers can tune parametersfor the
dynamic repair strategies (e.g., number ofparallel repairs, automated versus
manual repairs) to maintainthe promised availability SLA.
System Requirements:
Hardware Requirements:
 System : Pentium IV 2.4 GHz.
 Hard Disk : 40 GB.
 Floppy Drive : 44 Mb.
 Monitor : 15 VGA Colour.
 Ram : 512 Mb.
Software Requirements:
 Operating system : Windows XP/7.
 Coding Language : net, C#.net
 Tool : Visual Studio 2010