SlideShare a Scribd company logo
1 of 83
Download to read offline
SERVICE PERFORMANCE AND
ANALYSIS IN CLOUD COMPUTING
Presented by: Abdulaziz Almabrouk Altagawy
Course: EC636 Stochastic and Random Process4 Oct 2018
1
Information
๏ƒ˜ Introduction
๏ƒ˜ Cloud Types
๏ƒ˜ PROBLEM
๏ƒ˜ SOLUTION
๏ƒ˜ PREPARATIONS
๏ƒ˜ THE PERCENTILE OF
RESPONSE TIME
๏ƒ˜ A COMPUTER SERVICE
PERFORMANCE MODEL
๏ƒ˜ A NUMERICAL
VALIDATION
๏ƒ˜ CONCLUSIONS
๏ƒ˜ Book(s)
๏ƒ˜ PROBABILITY, MARKOV
CHAINS, QUEUES, AND
SIMULATION 2009-
William J. Stewart
๏ƒ˜ Probability, Statistics, and
Random Processes for
Electrical Engineering
Materials, 3rd ed, Alberto
Leon-Garcia
๏ƒ˜ Introduction to Probability
Models, 10th ed, Ross
4 Oct 2018
2
Introduction
4 Oct 2018
3
Introduction
4 Oct 2018
4
๏‚จ Cloud Computing is an information technology (IT)
paradigm that enables everywhere access to the
shared pools of configurable system resources and
higher-level services that can be rapidly
provisioned with minimal management effort, often
over the Internet. Cloud computing relies on sharing
of resources to achieve coherence and economies of
scale, similar to a public utility [Wikipedia].
Introduction
๏‚จ Examples of cloud service
include online file storage,
social networking sites,
webmail and online business
applications. The cloud
computing model allows
access to information and
computer resources from
anywhere that a network
connection is available. Cloud
Computing provides a shared
pool of resources, including
data storage space, networks,
computer processing power,
and specialized corporate
and user applications.
4 Oct 2018
5
Cloud Types
4 Oct 2018
6
Cloud Types
4 Oct 2018
7
๏‚จ Deployment models: This refers to the location and management of the cloud's
infrastructure.
๏ƒ˜ The NIST (The U.S. National Institute of Standards and Technology) definition for
the four deployment models is as follows:
๏ƒ˜ Public cloud: The public cloud infrastructure is available for public use
alternatively for a large industry group and is owned by an organization selling
cloud services.
๏ƒ˜ Private cloud: The private cloud infrastructure is operated for the exclusive use
of an organization. The cloud may be managed by that organization or a third
party. Private clouds may be either on- or off-premises.
๏ƒ˜ Hybrid cloud: A hybrid cloud combines multiple clouds (private, community of
public) where those clouds retain their unique identities, but are bound together
as a unit. A hybrid cloud may offer standardized or proprietary access to data
and applications, as well as application portability.
๏ƒ˜ Community cloud: A community cloud is one where the cloud has been
organized to serve a common function or purpose.
Cloud Types
4 Oct 2018
8
๏‚จ Service models: This consists of the particular types of services that you can access on a cloud computing
platform.
๏‚จ Three service types have been universally accepted:
๏ƒ˜ Infrastructure as a Service: IaaS provides virtual machines, virtual storage, virtual infrastructure, and
other hardware assets as resources that clients can provision.
๏ƒ˜ The IaaS service provider manages all the infrastructure, while the client is responsible for all other
aspects of the deployment. This can include the operating system, applications, and user interactions with
the system.
๏ƒ˜ Platform as a Service: PaaS provides virtual machines, operating systems, applications, services,
development frameworks, transactions, and control structures.
๏ƒ˜ The client can deploy its applications on the cloud infrastructure or use applications that were
programmed using languages and tools that are supported by the PaaS service provider. The service
provider manages the cloud infrastructure, the operating systems, and the enabling software. The client is
responsible for installing and managing the application that it is deploying.
๏ƒ˜ Software as a Service: SaaS is a complete operating environment with applications, management, and
the user interface.
๏ƒ˜ In the SaaS model, the application is provided to the client through a thin client interface (a browser,
usually), and the customer's responsibility begins and ends with entering and managing its data and user
interaction. Everything from the application down to the infrastructure is the vendor's responsibility.
Cloud Types
4 Oct 2018
9
PROBLEM
4 Oct 2018
10
PROBLEM
4 Oct 2018
11
๏‚จ Customers pay only for the used computer resources and
services by means of customized service level agreement
(SLA), also they have no knowledge of how a service
provider uses an underlying computer technological
infrastructure to support them.
๏‚จ The SLA is a contract negotiated and agreed between a
customer and a service provider. That is, the service
provider is required to execute service requests from a
customer within negotiated quality of service (QoS)
requirements for a given price. So,
based on system statistics
and a customerโ€™s perceived quality allows a service
provider to assure the quality of services and also avoid
over provisioning to meet an SLA.
PROBLEM
๏‚จ The majority of current
cloud computing
infrastructure as of
2009 consist of services
that are offered up and
delivered through a
service center such as a
data center that can be
accessed from a web
browser anywhere in the
world.
4 Oct 2018
12
PROBLEM
๏‚จ This model is a single
point of access for the
cloud computing needs
of the customers being
serviced through a Web
browser supported by a
Web server. The service
application running will
be according to the SLA
that customer pay for.
4 Oct 2018
13
PROBLEM
4 Oct 2018
14
๏‚จ the QoS requirements in computer service
performance usually uses the
๏‚จ This easy.
๏‚จ But it does not address the concerns of a customer.
๏‚จ The customer is preferred to request a
than an average
response time.
PROBLEM
4 Oct 2018
15
๏‚จ So our concern to find the of the
response time that characterizes the statistical
response time.
is the time of executing a service request
is less than a pre-defined value with a certain
percentage of time.
PROBLEM
๏‚จ There are three important but
challenging questions for customer
service performance in cloud
computing need to have an
answer for:
1) For a given arrival rate of
service requests and given
service rates at the Web server
and the service center, what
level of QoS services can be
guaranteed?
2) What are minimal service rates
required at the Web server
and the service center
respectively so that a given
percentile of the response time
can be guaranteed for a given
service arrival rate from
customers?
3) How many numbers of
customers can be supported so
that a given percentile of the
response time can be still
guaranteed when service rates
are given at the Web server
and the service center
respectively?
4 Oct 2018
16
SOLUTION
4 Oct 2018
17
SOLUTION
4 Oct 2018
18
๏‚จ So to compute a percentile of the response time one has first
to find the of the response time.
๏‚จ Walrand and Varaiya [23] showed that in any open Jackson
network, the response times of a customer at the various nodes
of overtake-free path are all mutually independent.
๏‚จ The goal to derive an approximation method for the
calculation of the probability and cumulative distributions of
the response time, and show the accuracy of the proposed
approximation method based on the obtained percentile
response time (or the cumulative distribution of response time),
and also to derive propositions and corollaries to answer the
previous mentioned service performance questions.
๏ƒ˜ Reversibility.
๏ƒ˜ Renewal Theory.
๏ƒ˜ LITTLEโ€™S FORMULA
๏ƒ˜ M/M/1 Queueing System.
โ€ข PASTA: Poisson Arrivals See Time Averages.
๏ƒ˜ Open Networks of Queues.
๏ƒ˜ Laplaceโ€“Stieltjes transforms.
๏ƒ˜ Arg max.
Preparations
4 Oct 2018
19
Reversibility
4 Oct 2018
20
Reversibility
4 Oct 2018
21
Reversibility
4 Oct 2018
22
Renewal Theory
4 Oct 2018
23
Renewal Theory
4 Oct 2018
24
Renewal Theory
4 Oct 2018
25
Renewal Theory
4 Oct 2018
26
Renewal Theory
4 Oct 2018
27
Renewal Theory
4 Oct 2018
28
Renewal Theory
4 Oct 2018
29
Renewal Theory
4 Oct 2018
30
LITTLEโ€™S FORMULA
4 Oct 2018
31
LITTLEโ€™S FORMULA
4 Oct 2018
32
LITTLEโ€™S FORMULA
4 Oct 2018
33
LITTLEโ€™S FORMULA
4 Oct 2018
34
LITTLEโ€™S FORMULA
4 Oct 2018
35
M/M/1 Queueing System
4 Oct 2018
36
M/M/1 Queueing System
4 Oct 2018
37
M/M/1 Queueing System
4 Oct 2018
38
M/M/1 Queueing System
4 Oct 2018
39
PASTA: Poisson Arrivals See Time
Averages
4 Oct 2018
40
PASTA: Poisson Arrivals See Time
Averages
4 Oct 2018
41
M/M/1 Queueing System
4 Oct 2018
42
Open Networks of Queues
4 Oct 2018
43
Open Networks of Queues
4 Oct 2018
44
Laplaceโ€“Stieltjes transforms
4 Oct 2018
45
Laplaceโ€“Stieltjes transforms
4 Oct 2018
46
Laplaceโ€“Stieltjes transforms
4 Oct 2018
47
Laplaceโ€“Stieltjes transforms
4 Oct 2018
48
Arg max
4 Oct 2018
49
THE PERCENTILE OF RESPONSE TIME
4 Oct 2018
50
THE PERCENTILE OF RESPONSE TIME
4 Oct 2018
51
๏‚จ Assume that ๐‘“๐‘‡ ๐‘ก be the probability distribution
function of a response time ๐‘‡ .
๏‚จ ๐‘‡ ๐ท
is a desired target response time that a
customer requests and agrees with its service
provider based on a fee paid by the customer.
๏‚จ The SLA performance metric that a ฮณ% SLA service
is guaranteed is as follows.
THE PERCENTILE OF RESPONSE TIME
4 Oct 2018
52
๏‚จ That is, ฮณ% of the time a customer will receive its
service in less than ๐‘‡ ๐ท
.
THE PERCENTILE OF RESPONSE TIME
4 Oct 2018
53
๏‚จ So in order to guarantee higher SLA service levels,
ฮผ increases when ๐‘‡ ๐ท
decreases. Similarly, for any
given arrival rate ฮป and service rate ฮผ , we can use
(2) to find the percentile of ฮณ . For example, when ฮป
= 100 and ๐‘‡ ๐ท
= 0.05.
THE PERCENTILE OF RESPONSE TIME
๏‚จ We can see that this service rate has to be bigger than 150
in order that 90% of the response time is less than 0.05.
4 Oct 2018
54
COMPUTER SERVICE
PERFORMANCE MODEL
4 Oct 2018
55
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
56
๏‚จ Computing The Response Time Distribution:
๏‚จ Modeling the customer service requests as a queueing network
model appears one of the best ways that makes it possible to not
only compute percentile response time but also characterize a
variable load in cloud computing.
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
57
๏‚จ We are going to derive the Laplace-Stieltjes transform
(LST) (simply called the Laplace transform alternatively)
of response time .
๏‚จ Let (๐‘– , ๐‘—) be the number of visits in the Web server
and the service center where ๐‘– and ๐‘— are the number of
visits in the Web server and the number of visits in the
service center respectively.
๏‚จ Let p(๐‘–; ๐‘—) be the probability of ๐‘– visits to the Web
server and ๐‘— visits to the service center. There may be
one time visit difference between the Web server and
the service center.
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
58
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
59
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
60
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
61
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
62
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
63
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
64
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
65
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
66
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
67
๏‚จ Proposition 1: For a given arrival rate ฮป service
rates ฮผ1 in the Web server and ฮผ2 in the service
center, and a set of parameters ฮฑ, ฮฒ and ๐‘‡ ๐ท
, the
level of QoS guaranteed services (ฮณ) will be no
more than 100๐น ๐‘‡ ๐‘‡ ๐ท
, ฮป, ฮผ1, ฮผ2 .
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
68
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
69
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
70
๏‚จ Now we interesting to find the closed-form
expression of ๐น ๐‘‡ ๐‘‡ ๐ท
, ฮป, ฮผ1, ฮผ2 .
๏‚จ Let ฮป๐‘– (๐‘– = 1; 2) be the arrival rates at the first and
second M/M/1 queues respectively.
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
71
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
72
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
73
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
74
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
75
A COMPUTER SERVICE PERFORMANCE
MODEL
4 Oct 2018
76
NUMERICAL VALIDATION
4 Oct 2018
77
NUMERICAL VALIDATION
4 Oct 2018
78
NUMERICAL VALIDATION
4 Oct 2018
79
Conclusion
4 Oct 2018
80
Conclusion
๏‚จ We have studied three important but challenging
questions for computer service performance in cloud
computing below: (1) For given service resources,
what level of QoS services can be guaranteed? (2)
For a given number of customers, how many service
resources are required to ensure that customer
services can be guaranteed in term of the
percentile of response time? (3) For given service
resources, how many customers can be supported to
ensure that customer services can be guaranteed in
term of the percentile of
4 Oct 2018
81
Conclusion
๏‚จ we have derived three propositions and three
corollaries for answering the above three questions.
The answers to the above three questions can be
obtained by using a numerical approximate method
in these propositions and corollaries.
4 Oct 2018
82
THANKS
4 Oct 2018
83

More Related Content

What's hot

A COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONS
A COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONSA COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONS
A COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONS
ijwscjournal
ย 
A latency-aware max-min algorithm for resource allocation in cloud
A latency-aware max-min algorithm for resource  allocation in cloud A latency-aware max-min algorithm for resource  allocation in cloud
A latency-aware max-min algorithm for resource allocation in cloud
IJECEIAES
ย 

What's hot (17)

DESIGNING ASPECT AND FUNCTIONALITY ISSUES OF CLOUD BROKERING SERVICE IN CLOUD...
DESIGNING ASPECT AND FUNCTIONALITY ISSUES OF CLOUD BROKERING SERVICE IN CLOUD...DESIGNING ASPECT AND FUNCTIONALITY ISSUES OF CLOUD BROKERING SERVICE IN CLOUD...
DESIGNING ASPECT AND FUNCTIONALITY ISSUES OF CLOUD BROKERING SERVICE IN CLOUD...
ย 
IRJET- Trust Value Calculation for Cloud Resources
IRJET- Trust Value Calculation for Cloud ResourcesIRJET- Trust Value Calculation for Cloud Resources
IRJET- Trust Value Calculation for Cloud Resources
ย 
Mmb authenticated index for verifiable
Mmb authenticated index for verifiableMmb authenticated index for verifiable
Mmb authenticated index for verifiable
ย 
A profit maximization scheme with guaranteed quality of service in cloud comp...
A profit maximization scheme with guaranteed quality of service in cloud comp...A profit maximization scheme with guaranteed quality of service in cloud comp...
A profit maximization scheme with guaranteed quality of service in cloud comp...
ย 
A PROFIT MAXIMIZATION SCHEME WITH GUARANTEED QUALITY OF SERVICE IN CLOUD COMP...
A PROFIT MAXIMIZATION SCHEME WITH GUARANTEED QUALITY OF SERVICE IN CLOUD COMP...A PROFIT MAXIMIZATION SCHEME WITH GUARANTEED QUALITY OF SERVICE IN CLOUD COMP...
A PROFIT MAXIMIZATION SCHEME WITH GUARANTEED QUALITY OF SERVICE IN CLOUD COMP...
ย 
A survey on various resource allocation policies in cloud computing environment
A survey on various resource allocation policies in cloud computing environmentA survey on various resource allocation policies in cloud computing environment
A survey on various resource allocation policies in cloud computing environment
ย 
50120130406006
5012013040600650120130406006
50120130406006
ย 
A profit maximization scheme with guaranteed quality of service in cloud comp...
A profit maximization scheme with guaranteed quality of service in cloud comp...A profit maximization scheme with guaranteed quality of service in cloud comp...
A profit maximization scheme with guaranteed quality of service in cloud comp...
ย 
A cloud broker approach with qos attendance and soa for hybrid cloud computin...
A cloud broker approach with qos attendance and soa for hybrid cloud computin...A cloud broker approach with qos attendance and soa for hybrid cloud computin...
A cloud broker approach with qos attendance and soa for hybrid cloud computin...
ย 
BEST FINAL YEAR PROJECT IEEE 2015 BY SPECTRUM SOLUTIONS PONDICHERRY
BEST FINAL YEAR PROJECT IEEE 2015 BY SPECTRUM SOLUTIONS PONDICHERRYBEST FINAL YEAR PROJECT IEEE 2015 BY SPECTRUM SOLUTIONS PONDICHERRY
BEST FINAL YEAR PROJECT IEEE 2015 BY SPECTRUM SOLUTIONS PONDICHERRY
ย 
CPET- Project Report
CPET- Project ReportCPET- Project Report
CPET- Project Report
ย 
Resource usage optimization in cloud based networks
Resource usage optimization in cloud based networksResource usage optimization in cloud based networks
Resource usage optimization in cloud based networks
ย 
A COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONS
A COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONSA COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONS
A COMPREHENSIVE SOLUTION TO CLOUD TRAFFIC TRIBULATIONS
ย 
ACS Seminar: Components & perceptions of SerVal in B2B cloud computing
ACS Seminar: Components & perceptions of SerVal in B2B cloud computingACS Seminar: Components & perceptions of SerVal in B2B cloud computing
ACS Seminar: Components & perceptions of SerVal in B2B cloud computing
ย 
Parallel and Distributed System IEEE 2015 Projects
Parallel and Distributed System IEEE 2015 ProjectsParallel and Distributed System IEEE 2015 Projects
Parallel and Distributed System IEEE 2015 Projects
ย 
IEEE 2015-2016 A Profit Maximization Scheme with Guaranteed Quality of Servic...
IEEE 2015-2016 A Profit Maximization Scheme with Guaranteed Quality of Servic...IEEE 2015-2016 A Profit Maximization Scheme with Guaranteed Quality of Servic...
IEEE 2015-2016 A Profit Maximization Scheme with Guaranteed Quality of Servic...
ย 
A latency-aware max-min algorithm for resource allocation in cloud
A latency-aware max-min algorithm for resource  allocation in cloud A latency-aware max-min algorithm for resource  allocation in cloud
A latency-aware max-min algorithm for resource allocation in cloud
ย 

Similar to Service performance and analysis in cloud computing extened 2

Performance and Cost Analysis of Modern Public Cloud Services
Performance and Cost Analysis of Modern Public Cloud ServicesPerformance and Cost Analysis of Modern Public Cloud Services
Performance and Cost Analysis of Modern Public Cloud Services
Md.Saiedur Rahaman
ย 
A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...
A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...
A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...
1crore projects
ย 
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
ijccsa
ย 
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
ijccsa
ย 
Strategic Business Challenges in Cloud Systems
Strategic Business Challenges in Cloud SystemsStrategic Business Challenges in Cloud Systems
Strategic Business Challenges in Cloud Systems
neirew J
ย 
Strategic business challenges in cloud
Strategic business challenges in cloudStrategic business challenges in cloud
Strategic business challenges in cloud
ijccsa
ย 
Cloud provider transparency
Cloud provider transparencyCloud provider transparency
Cloud provider transparency
Prachyanun Nilsook
ย 

Similar to Service performance and analysis in cloud computing extened 2 (20)

A Study On Service Level Agreement Management Techniques In Cloud
A Study On Service Level Agreement Management Techniques In CloudA Study On Service Level Agreement Management Techniques In Cloud
A Study On Service Level Agreement Management Techniques In Cloud
ย 
Revenue Maximization with Good Quality of Service in Cloud Computing
Revenue Maximization with Good Quality of Service in Cloud ComputingRevenue Maximization with Good Quality of Service in Cloud Computing
Revenue Maximization with Good Quality of Service in Cloud Computing
ย 
ADAPTIVE MULTI-TENANCY POLICY FOR ENHANCING SERVICE LEVEL AGREEMENT THROUGH R...
ADAPTIVE MULTI-TENANCY POLICY FOR ENHANCING SERVICE LEVEL AGREEMENT THROUGH R...ADAPTIVE MULTI-TENANCY POLICY FOR ENHANCING SERVICE LEVEL AGREEMENT THROUGH R...
ADAPTIVE MULTI-TENANCY POLICY FOR ENHANCING SERVICE LEVEL AGREEMENT THROUGH R...
ย 
Performance and Cost Analysis of Modern Public Cloud Services
Performance and Cost Analysis of Modern Public Cloud ServicesPerformance and Cost Analysis of Modern Public Cloud Services
Performance and Cost Analysis of Modern Public Cloud Services
ย 
Pricing Models for Cloud Computing Services, a Survey
Pricing Models for Cloud Computing Services, a SurveyPricing Models for Cloud Computing Services, a Survey
Pricing Models for Cloud Computing Services, a Survey
ย 
A survey on various resource allocation policies in cloud computing environment
A survey on various resource allocation policies in cloud computing environmentA survey on various resource allocation policies in cloud computing environment
A survey on various resource allocation policies in cloud computing environment
ย 
A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...
A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...
A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Comp...
ย 
Dynamic congestion management system for cloud service broker
Dynamic congestion management system for cloud service  brokerDynamic congestion management system for cloud service  broker
Dynamic congestion management system for cloud service broker
ย 
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
ย 
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
THE EFFECT OF THE RESOURCE CONSUMPTION CHARACTERISTICS OF CLOUD APPLICATIONS ...
ย 
Profit Maximization for Service Providers using Hybrid Pricing in Cloud Compu...
Profit Maximization for Service Providers using Hybrid Pricing in Cloud Compu...Profit Maximization for Service Providers using Hybrid Pricing in Cloud Compu...
Profit Maximization for Service Providers using Hybrid Pricing in Cloud Compu...
ย 
Strategic Business Challenges in Cloud Systems
Strategic Business Challenges in Cloud SystemsStrategic Business Challenges in Cloud Systems
Strategic Business Challenges in Cloud Systems
ย 
Strategic business challenges in cloud
Strategic business challenges in cloudStrategic business challenges in cloud
Strategic business challenges in cloud
ย 
Pricing model
Pricing modelPricing model
Pricing model
ย 
A SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTING
A SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTINGA SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTING
A SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTING
ย 
A SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTING
A SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTINGA SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTING
A SURVEY ON RESOURCE ALLOCATION IN CLOUD COMPUTING
ย 
A Survey on Resource Allocation in Cloud Computing
A Survey on Resource Allocation in Cloud ComputingA Survey on Resource Allocation in Cloud Computing
A Survey on Resource Allocation in Cloud Computing
ย 
Cloud provider transparency
Cloud provider transparencyCloud provider transparency
Cloud provider transparency
ย 
20150113
2015011320150113
20150113
ย 
Cloud computing for java and dotnet
Cloud computing for java and dotnetCloud computing for java and dotnet
Cloud computing for java and dotnet
ย 

More from Abdullaziz Tagawy

EMP_Assessment Report ABDELAZEZ TAGAWY
EMP_Assessment Report ABDELAZEZ TAGAWYEMP_Assessment Report ABDELAZEZ TAGAWY
EMP_Assessment Report ABDELAZEZ TAGAWY
Abdullaziz Tagawy
ย 
Error Detection and Correction - Data link Layer
Error Detection and Correction - Data link LayerError Detection and Correction - Data link Layer
Error Detection and Correction - Data link Layer
Abdullaziz Tagawy
ย 
Introduction to Data-Link Layer
Introduction to Data-Link LayerIntroduction to Data-Link Layer
Introduction to Data-Link Layer
Abdullaziz Tagawy
ย 

More from Abdullaziz Tagawy (13)

Solar Cells versus Photodiode
Solar Cells versus PhotodiodeSolar Cells versus Photodiode
Solar Cells versus Photodiode
ย 
Managing enterprise networks with cisco prime infrastructure_ 1 of 2
Managing enterprise networks with cisco prime infrastructure_ 1 of 2Managing enterprise networks with cisco prime infrastructure_ 1 of 2
Managing enterprise networks with cisco prime infrastructure_ 1 of 2
ย 
EMP_Assessment Report ABDELAZEZ TAGAWY
EMP_Assessment Report ABDELAZEZ TAGAWYEMP_Assessment Report ABDELAZEZ TAGAWY
EMP_Assessment Report ABDELAZEZ TAGAWY
ย 
IPSec and VPN
IPSec and VPNIPSec and VPN
IPSec and VPN
ย 
Convolutional codes
Convolutional codesConvolutional codes
Convolutional codes
ย 
OFDM Orthogonal Frequency Division Multiplexing
OFDM Orthogonal Frequency Division MultiplexingOFDM Orthogonal Frequency Division Multiplexing
OFDM Orthogonal Frequency Division Multiplexing
ย 
Solving QoS multicast routing problem using ACO algorithm
Solving QoS multicast routing problem using ACO algorithmSolving QoS multicast routing problem using ACO algorithm
Solving QoS multicast routing problem using ACO algorithm
ย 
Solving QoS multicast routing problem using aco algorithm
Solving QoS multicast routing problem using aco algorithm Solving QoS multicast routing problem using aco algorithm
Solving QoS multicast routing problem using aco algorithm
ย 
SNAPDRAGON SoC Family and ARM Architecture
SNAPDRAGON SoC Family and ARM Architecture SNAPDRAGON SoC Family and ARM Architecture
SNAPDRAGON SoC Family and ARM Architecture
ย 
Error Detection and Correction - Data link Layer
Error Detection and Correction - Data link LayerError Detection and Correction - Data link Layer
Error Detection and Correction - Data link Layer
ย 
Introduction to Data-Link Layer
Introduction to Data-Link LayerIntroduction to Data-Link Layer
Introduction to Data-Link Layer
ย 
Comp net 2
Comp net 2Comp net 2
Comp net 2
ย 
Comp net 1
Comp net 1Comp net 1
Comp net 1
ย 

Recently uploaded

โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...
โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...
โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...
Diya Sharma
ย 
( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...
( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...
( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...
nilamkumrai
ย 
VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...
SUHANI PANDEY
ย 
Call Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort ServiceCall Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
ย 
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
SUHANI PANDEY
ย 
Lucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRL
Lucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRLLucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRL
Lucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRL
imonikaupta
ย 
Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...
Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...
Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...
SUHANI PANDEY
ย 
VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...
VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...
VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...
SUHANI PANDEY
ย 
Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...
Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...
Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...
SUHANI PANDEY
ย 

Recently uploaded (20)

(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
ย 
โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...
โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...
โ‚น5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] ๐Ÿ”|97111...
ย 
( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...
( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...
( Pune ) VIP Baner Call Girls ๐ŸŽ—๏ธ 9352988975 Sizzling | Escorts | Girls Are Re...
ย 
VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Mohammadwadi WhatSapp Number 8005736733 With Elite Staff...
ย 
Call Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort ServiceCall Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Prashant Vihar, Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
ย 
Call Now โ˜Ž 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
Call Now โ˜Ž 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.Call Now โ˜Ž 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
Call Now โ˜Ž 8264348440 !! Call Girls in Shahpur Jat Escort Service Delhi N.C.R.
ย 
Trump Diapers Over Dems t shirts Sweatshirt
Trump Diapers Over Dems t shirts SweatshirtTrump Diapers Over Dems t shirts Sweatshirt
Trump Diapers Over Dems t shirts Sweatshirt
ย 
Real Men Wear Diapers T Shirts sweatshirt
Real Men Wear Diapers T Shirts sweatshirtReal Men Wear Diapers T Shirts sweatshirt
Real Men Wear Diapers T Shirts sweatshirt
ย 
Pune Airport ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready...
Pune Airport ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready...Pune Airport ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready...
Pune Airport ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready...
ย 
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
ย 
Lucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRL
Lucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRLLucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRL
Lucknow โคCALL GIRL 88759*99948 โคCALL GIRLS IN Lucknow ESCORT SERVICEโคCALL GIRL
ย 
Nanded City ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready ...
Nanded City ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready ...Nanded City ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready ...
Nanded City ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready ...
ย 
APNIC Updates presented by Paul Wilson at ARIN 53
APNIC Updates presented by Paul Wilson at ARIN 53APNIC Updates presented by Paul Wilson at ARIN 53
APNIC Updates presented by Paul Wilson at ARIN 53
ย 
Call Now โ˜Ž 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.
Call Now โ˜Ž 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.Call Now โ˜Ž 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.
Call Now โ˜Ž 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.
ย 
Russian Call Girls in %(+971524965298 )# Call Girls in Dubai
Russian Call Girls in %(+971524965298  )#  Call Girls in DubaiRussian Call Girls in %(+971524965298  )#  Call Girls in Dubai
Russian Call Girls in %(+971524965298 )# Call Girls in Dubai
ย 
Dubai Call Girls Milky O525547819 Call Girls Dubai Soft Dating
Dubai Call Girls Milky O525547819 Call Girls Dubai Soft DatingDubai Call Girls Milky O525547819 Call Girls Dubai Soft Dating
Dubai Call Girls Milky O525547819 Call Girls Dubai Soft Dating
ย 
Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...
Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...
Sarola * Female Escorts Service in Pune | 8005736733 Independent Escorts & Da...
ย 
VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...
VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...
VVIP Pune Call Girls Sinhagad WhatSapp Number 8005736733 With Elite Staff And...
ย 
Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...
Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...
Wagholi & High Class Call Girls Pune Neha 8005736733 | 100% Gennuine High Cla...
ย 
Al Barsha Night Partner +0567686026 Call Girls Dubai
Al Barsha Night Partner +0567686026 Call Girls  DubaiAl Barsha Night Partner +0567686026 Call Girls  Dubai
Al Barsha Night Partner +0567686026 Call Girls Dubai
ย 

Service performance and analysis in cloud computing extened 2

  • 1. SERVICE PERFORMANCE AND ANALYSIS IN CLOUD COMPUTING Presented by: Abdulaziz Almabrouk Altagawy Course: EC636 Stochastic and Random Process4 Oct 2018 1
  • 2. Information ๏ƒ˜ Introduction ๏ƒ˜ Cloud Types ๏ƒ˜ PROBLEM ๏ƒ˜ SOLUTION ๏ƒ˜ PREPARATIONS ๏ƒ˜ THE PERCENTILE OF RESPONSE TIME ๏ƒ˜ A COMPUTER SERVICE PERFORMANCE MODEL ๏ƒ˜ A NUMERICAL VALIDATION ๏ƒ˜ CONCLUSIONS ๏ƒ˜ Book(s) ๏ƒ˜ PROBABILITY, MARKOV CHAINS, QUEUES, AND SIMULATION 2009- William J. Stewart ๏ƒ˜ Probability, Statistics, and Random Processes for Electrical Engineering Materials, 3rd ed, Alberto Leon-Garcia ๏ƒ˜ Introduction to Probability Models, 10th ed, Ross 4 Oct 2018 2
  • 4. Introduction 4 Oct 2018 4 ๏‚จ Cloud Computing is an information technology (IT) paradigm that enables everywhere access to the shared pools of configurable system resources and higher-level services that can be rapidly provisioned with minimal management effort, often over the Internet. Cloud computing relies on sharing of resources to achieve coherence and economies of scale, similar to a public utility [Wikipedia].
  • 5. Introduction ๏‚จ Examples of cloud service include online file storage, social networking sites, webmail and online business applications. The cloud computing model allows access to information and computer resources from anywhere that a network connection is available. Cloud Computing provides a shared pool of resources, including data storage space, networks, computer processing power, and specialized corporate and user applications. 4 Oct 2018 5
  • 7. Cloud Types 4 Oct 2018 7 ๏‚จ Deployment models: This refers to the location and management of the cloud's infrastructure. ๏ƒ˜ The NIST (The U.S. National Institute of Standards and Technology) definition for the four deployment models is as follows: ๏ƒ˜ Public cloud: The public cloud infrastructure is available for public use alternatively for a large industry group and is owned by an organization selling cloud services. ๏ƒ˜ Private cloud: The private cloud infrastructure is operated for the exclusive use of an organization. The cloud may be managed by that organization or a third party. Private clouds may be either on- or off-premises. ๏ƒ˜ Hybrid cloud: A hybrid cloud combines multiple clouds (private, community of public) where those clouds retain their unique identities, but are bound together as a unit. A hybrid cloud may offer standardized or proprietary access to data and applications, as well as application portability. ๏ƒ˜ Community cloud: A community cloud is one where the cloud has been organized to serve a common function or purpose.
  • 8. Cloud Types 4 Oct 2018 8 ๏‚จ Service models: This consists of the particular types of services that you can access on a cloud computing platform. ๏‚จ Three service types have been universally accepted: ๏ƒ˜ Infrastructure as a Service: IaaS provides virtual machines, virtual storage, virtual infrastructure, and other hardware assets as resources that clients can provision. ๏ƒ˜ The IaaS service provider manages all the infrastructure, while the client is responsible for all other aspects of the deployment. This can include the operating system, applications, and user interactions with the system. ๏ƒ˜ Platform as a Service: PaaS provides virtual machines, operating systems, applications, services, development frameworks, transactions, and control structures. ๏ƒ˜ The client can deploy its applications on the cloud infrastructure or use applications that were programmed using languages and tools that are supported by the PaaS service provider. The service provider manages the cloud infrastructure, the operating systems, and the enabling software. The client is responsible for installing and managing the application that it is deploying. ๏ƒ˜ Software as a Service: SaaS is a complete operating environment with applications, management, and the user interface. ๏ƒ˜ In the SaaS model, the application is provided to the client through a thin client interface (a browser, usually), and the customer's responsibility begins and ends with entering and managing its data and user interaction. Everything from the application down to the infrastructure is the vendor's responsibility.
  • 11. PROBLEM 4 Oct 2018 11 ๏‚จ Customers pay only for the used computer resources and services by means of customized service level agreement (SLA), also they have no knowledge of how a service provider uses an underlying computer technological infrastructure to support them. ๏‚จ The SLA is a contract negotiated and agreed between a customer and a service provider. That is, the service provider is required to execute service requests from a customer within negotiated quality of service (QoS) requirements for a given price. So, based on system statistics and a customerโ€™s perceived quality allows a service provider to assure the quality of services and also avoid over provisioning to meet an SLA.
  • 12. PROBLEM ๏‚จ The majority of current cloud computing infrastructure as of 2009 consist of services that are offered up and delivered through a service center such as a data center that can be accessed from a web browser anywhere in the world. 4 Oct 2018 12
  • 13. PROBLEM ๏‚จ This model is a single point of access for the cloud computing needs of the customers being serviced through a Web browser supported by a Web server. The service application running will be according to the SLA that customer pay for. 4 Oct 2018 13
  • 14. PROBLEM 4 Oct 2018 14 ๏‚จ the QoS requirements in computer service performance usually uses the ๏‚จ This easy. ๏‚จ But it does not address the concerns of a customer. ๏‚จ The customer is preferred to request a than an average response time.
  • 15. PROBLEM 4 Oct 2018 15 ๏‚จ So our concern to find the of the response time that characterizes the statistical response time. is the time of executing a service request is less than a pre-defined value with a certain percentage of time.
  • 16. PROBLEM ๏‚จ There are three important but challenging questions for customer service performance in cloud computing need to have an answer for: 1) For a given arrival rate of service requests and given service rates at the Web server and the service center, what level of QoS services can be guaranteed? 2) What are minimal service rates required at the Web server and the service center respectively so that a given percentile of the response time can be guaranteed for a given service arrival rate from customers? 3) How many numbers of customers can be supported so that a given percentile of the response time can be still guaranteed when service rates are given at the Web server and the service center respectively? 4 Oct 2018 16
  • 18. SOLUTION 4 Oct 2018 18 ๏‚จ So to compute a percentile of the response time one has first to find the of the response time. ๏‚จ Walrand and Varaiya [23] showed that in any open Jackson network, the response times of a customer at the various nodes of overtake-free path are all mutually independent. ๏‚จ The goal to derive an approximation method for the calculation of the probability and cumulative distributions of the response time, and show the accuracy of the proposed approximation method based on the obtained percentile response time (or the cumulative distribution of response time), and also to derive propositions and corollaries to answer the previous mentioned service performance questions.
  • 19. ๏ƒ˜ Reversibility. ๏ƒ˜ Renewal Theory. ๏ƒ˜ LITTLEโ€™S FORMULA ๏ƒ˜ M/M/1 Queueing System. โ€ข PASTA: Poisson Arrivals See Time Averages. ๏ƒ˜ Open Networks of Queues. ๏ƒ˜ Laplaceโ€“Stieltjes transforms. ๏ƒ˜ Arg max. Preparations 4 Oct 2018 19
  • 36. M/M/1 Queueing System 4 Oct 2018 36
  • 37. M/M/1 Queueing System 4 Oct 2018 37
  • 38. M/M/1 Queueing System 4 Oct 2018 38
  • 39. M/M/1 Queueing System 4 Oct 2018 39
  • 40. PASTA: Poisson Arrivals See Time Averages 4 Oct 2018 40
  • 41. PASTA: Poisson Arrivals See Time Averages 4 Oct 2018 41
  • 42. M/M/1 Queueing System 4 Oct 2018 42
  • 43. Open Networks of Queues 4 Oct 2018 43
  • 44. Open Networks of Queues 4 Oct 2018 44
  • 49. Arg max 4 Oct 2018 49
  • 50. THE PERCENTILE OF RESPONSE TIME 4 Oct 2018 50
  • 51. THE PERCENTILE OF RESPONSE TIME 4 Oct 2018 51 ๏‚จ Assume that ๐‘“๐‘‡ ๐‘ก be the probability distribution function of a response time ๐‘‡ . ๏‚จ ๐‘‡ ๐ท is a desired target response time that a customer requests and agrees with its service provider based on a fee paid by the customer. ๏‚จ The SLA performance metric that a ฮณ% SLA service is guaranteed is as follows.
  • 52. THE PERCENTILE OF RESPONSE TIME 4 Oct 2018 52 ๏‚จ That is, ฮณ% of the time a customer will receive its service in less than ๐‘‡ ๐ท .
  • 53. THE PERCENTILE OF RESPONSE TIME 4 Oct 2018 53 ๏‚จ So in order to guarantee higher SLA service levels, ฮผ increases when ๐‘‡ ๐ท decreases. Similarly, for any given arrival rate ฮป and service rate ฮผ , we can use (2) to find the percentile of ฮณ . For example, when ฮป = 100 and ๐‘‡ ๐ท = 0.05.
  • 54. THE PERCENTILE OF RESPONSE TIME ๏‚จ We can see that this service rate has to be bigger than 150 in order that 90% of the response time is less than 0.05. 4 Oct 2018 54
  • 56. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 56 ๏‚จ Computing The Response Time Distribution: ๏‚จ Modeling the customer service requests as a queueing network model appears one of the best ways that makes it possible to not only compute percentile response time but also characterize a variable load in cloud computing.
  • 57. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 57 ๏‚จ We are going to derive the Laplace-Stieltjes transform (LST) (simply called the Laplace transform alternatively) of response time . ๏‚จ Let (๐‘– , ๐‘—) be the number of visits in the Web server and the service center where ๐‘– and ๐‘— are the number of visits in the Web server and the number of visits in the service center respectively. ๏‚จ Let p(๐‘–; ๐‘—) be the probability of ๐‘– visits to the Web server and ๐‘— visits to the service center. There may be one time visit difference between the Web server and the service center.
  • 58. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 58
  • 59. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 59
  • 60. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 60
  • 61. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 61
  • 62. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 62
  • 63. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 63
  • 64. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 64
  • 65. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 65
  • 66. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 66
  • 67. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 67 ๏‚จ Proposition 1: For a given arrival rate ฮป service rates ฮผ1 in the Web server and ฮผ2 in the service center, and a set of parameters ฮฑ, ฮฒ and ๐‘‡ ๐ท , the level of QoS guaranteed services (ฮณ) will be no more than 100๐น ๐‘‡ ๐‘‡ ๐ท , ฮป, ฮผ1, ฮผ2 .
  • 68. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 68
  • 69. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 69
  • 70. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 70 ๏‚จ Now we interesting to find the closed-form expression of ๐น ๐‘‡ ๐‘‡ ๐ท , ฮป, ฮผ1, ฮผ2 . ๏‚จ Let ฮป๐‘– (๐‘– = 1; 2) be the arrival rates at the first and second M/M/1 queues respectively.
  • 71. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 71
  • 72. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 72
  • 73. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 73
  • 74. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 74
  • 75. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 75
  • 76. A COMPUTER SERVICE PERFORMANCE MODEL 4 Oct 2018 76
  • 81. Conclusion ๏‚จ We have studied three important but challenging questions for computer service performance in cloud computing below: (1) For given service resources, what level of QoS services can be guaranteed? (2) For a given number of customers, how many service resources are required to ensure that customer services can be guaranteed in term of the percentile of response time? (3) For given service resources, how many customers can be supported to ensure that customer services can be guaranteed in term of the percentile of 4 Oct 2018 81
  • 82. Conclusion ๏‚จ we have derived three propositions and three corollaries for answering the above three questions. The answers to the above three questions can be obtained by using a numerical approximate method in these propositions and corollaries. 4 Oct 2018 82