Performance and Cost Analysis of Modern Public Cloud Services

Md. Saiedur Rahaman
rsaied@yahoo.com
Supervisor: Vasilios Andrikopoulos
Performance and Cost Analysis of
Modern Public Cloud Services.

2© MD. SAIEDUR RAHAMAN
Outline
 Problem Statement.
 Performance and cost of public cloud.
 Overview.
 Performance analysis of public cloud.
 Cost analysis of public cloud.
 Existing Approaches & Comparison.
 For better performance of cloud.
 Approaches.
 Comparison.
 Summery.
 Discussion.
 Open Issues.
 Summery.
 Q&A.

Problem Statement
 Selection of public cloud providers based on performance & cost
 Determining public cloud cost: Different provider has differnt costing
approaces due to their structure and services.
 VM migration of public cloud: Most provider has on premises VMs migrate
to public clouds, some providers provide a graphical interface and some are
taxtual.
 VM template: While any public cloud provider allows you to create generic
VMs based on predefined templates, the complexity and number of those
templates varies wildly.
 Auto scalling: The basic idea of auto scaling is that server workloads are
linear. Sometimes, demands are peak, such as insurance companies' open
enrolment periods or online retailers' big holiday sales.

Problem Statement
 Network connectivity: Every cloud service provider offers VM network
connectivity. It is clear that your VMs will be able to access one another
through Internet. the larger cloud providers typically offer multiple network
connectivity options.
 Storage choice: Storage services vary tremendously from one provider to
another, but at a minimum, there are usually standard and premium options.
 Regional support: Pay attention to cloud providers' regional availability.
Larger cloud providers establish data centers all over the world. If you have
business requirements that mandate your data to keep within a specific
country, it's extremely important to be able to choose which data centers will
host your VMs.
 Operating System: We can consider the provider with different operating
system Linux and windows. We can say for booting instance and execution
Linux is faster than Windows.

Public Cloud computing
 Service model[**]:

Cost Model [15]
Approaches Fairness Pros Cons
Pay as you go(
set by
provider)
dynamic
Fair for
client, he
might not
pay for more
than need
Clients is aware of
the exact cost to be
paid and resources
are reserved for
clients for the time
periods he paid
1.service provider may reserve
the resources for longer time than
clients utilized
2. Service provider cannot raise
cost when demand is high;
Subscription
(Static)
Clients may
sometime
overpay or
underpay
Clients may
underpay for the
resources when he
use extremely
Clients may overpay for
resources when do not more use.
Pay for
resource
model-static
Fair for both
provider and
client
Offers maximum
utilization of the
service provider’s
resources
Hard to implement
Based initial
cost-dynamic
Not fair
to client
Simple to
calculate cost
Ignore client opinion

Cost Model
Approaches Fairness Pros Cons
Competition based
costing-dynamic
Fair to client Easy
to implement
Does not take consider
into client
Client based
costing(wish to pay)-
dynamic
Fair to client Client taken into
account
Client rarely indicate to
seller what they willing to
pay
Hybrid
costing(according
wait time)-static +
dynamic
Fair to client, it
automatically adjust
cost within static
limit
Simple and low
computational
overhead
Need to reach an
agreement
Costing algorithm
for cloud computing
resources-dynamic
Fair for provider
because it reduce
cost and maximized
revenue
reduce cost and
maximized
revenue
Model is almost fixed and
cannot rapid change of
supply on demand.

Performance of public cloud
 Performance of different cloud[23]
 Elasticity of different cloud[23]
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
Get Put Query
Cloud 1
Cloud 2
Cloud 3
0
10
20
30
40
50
60
Cloud 1 Cloud 2 Cloud 3
Used Instances
Reserted Instances
Total Intances

Performance of public cloud
 Upload and download speed of different cloud [25]
 Scaling latency for cloud with different OS[23]
0
0.5
1
1.5
2
2.5
Linux Task1 Windows Task1 Linux Task2 Windows Task2
Execution latency Booting Latency Total Latency
Cloud name Upload Download Time
C1 128Mbps 512 Mbps Off
128 Mbps 256 Mbps Peak
C2 256Mbps 512 Mbps Off
128 Mbps 512 Mbps Peak
C3 256Mbps 256 Mbps off
256Mbps 256 Mbps Peak

Performance Comparison of cloud [***]
Plateform Elastic
Cluster
Storage WideArea
Network
others
Amazon AWS Xen VM
(virtual machine
monitor)
SimpleDB(table),
S3(blob),
SQS(queue)
3 data center
(2 USA,
1 Europ)
Flexible,
proper load
balance, more
functions
Microsoft
Azure
Azure VM Xstore
(Table,blob,queue)
6 data center
(2 USA, 2
Europ,
2 Asia)
Running only
MS service &
technology
Google
AppEngine
Proprietary
Sandbox
(isolated
computing
environment )
DataStore(table) Unpublished
number of
Google
data center
Better network,
secure, faster
Rackspace
CloudServer
Xen VM CloudFiles(blob) 2 data
center (USA)
Better
ferformance,
cost,security

Cost of Public Cloud computing [15]
 Cost model: Cost is a critical factor for providers offering services or
products . Developing an appropriate cost model will help achieve higher
revenue for organization. It includes the manufacturing cost and maintenance
cost, market cost. Customer consider three parameters to choose provider.
 Costing approach: Costing approaches determine cost of service. it is [14].
• Fixed cost regardless volume.
• Fixed cost plus per unit rate.
 Assured purchase volume plus per unit cost rate.
 Per unit rate with ceiling (customer pay for certain limit of use).
 Per unit cost (customer pay different cost per unit).
 Quality of Service: If QoS is ensure by the provider, it will increase the
number of customer and loyalty of the service provider.
 Utilization period: Utilization periods reflect that, how long time a
customer can be used the service according to SLAs between customer &
provider.

Cost of Public Cloud computing
 Aspect of cloud computing [15]

Existing Approaches
 For better performance of cloud: It shold have some quality
parameters.
 High processing speed.
 High memory capacity.
 High bandwidth.
 Large storage capacity.
 Skilled personnel.
 Approaches: Different service provider offer different schemes and model
for costing.
 Amazon consider the market leader in cloud computing [16]. Amazon uses a
fixed cost model for each hour of VM usage.
 GAE and Microsoft Azure is also leader in cloud market use “pay as you go”
and “pay for resources” cost model.
 Subscriptions model where customer pays in advance for the services he is
going to receive for a predefine period of time.

Approach
 Overview SMI: SMI attributes designed based on International Standard
Organization (ISO) standard by the CSMIC(cloud service measurement
initiative consortium) consortium [2]. It is a set of business related Key
performance Indicators (KPI) which provides standard method for measuring
and comparing business service performance [26].
 Performance analysis of public cloud: Measurement parameters.
 Service response time.
 Sustainability.
 Accuracy.
 Adaptability.
 Elasticity.
 Throughput.
 Scaling latency.
 Internet bandwidth.

Approach
 Overview of CloudCMP Compare: Compared to traditional
computing model that uses in-house infrastructure, cloud computing offers
advantages of cost and reliability like no need infrastructure and maintenance
cost [23].
 Goals and approaches :
 Guided a client to select a cloud provider.
 Relevant to the cloud provider service.
 Fair.
 Thoroughness vs mesurement cost.
 Coverage vs development cost.

Approach
 Overview To move or not to move: Cloud computing give some
advantages as easy of management, infrastructure saving, salary saving.
Although many advantages some client will not get advantages from cloud
providers if they have already infrastructure and administrative base [27].
 Identify set of cost factor:
 Net present value.
• Workload intensity and growth.
• Data transfer.
• Storage capacity.
• Softwase license.
• Workload variance and cloud elasticity.
 Cost components.
 Application hosting choice.

Approach
 Overview performance & cost assessment: NICTA has developed
service oriented performance modeling(SOPM), a technology of service
oriented architecture(SOA) application for performance and scalability. SOA
has a sub task called cloud architecture. SOPM are parameterized by workload,
performance [25].
 Resource and cost.
 In-house hosting, bare hardware.
 In-house hosting, virtualization.
 Server performance variability.
 LAN latency and bandwidth.
 WAN latency and bandwidth.

Approach
 Overview performance analysis of cloud computing:There are
several solutions are offered by service provider for same problem addressing
to IT needs of different clients. All solution has different performance in term
of functionality, accuracy, service response time. Clients need to understand
how their application will perform in different cloud platform and where there
demand will be fulfilled or not [24].
 Cloud computing services for scientific computing: Job strcture & source,
Bottleneck resources, Job parallelism.
 Four Selected Clouds: GAE, AWS, Azure, Rackspace:
 MTC(many task cmp) presence in scientific computing workloads:
 Cloud performance evaluation: Multi machine benchmark, single machine
benchmark, resource book & release, performance stability.
 Clouds vs. others scientific computing infrastructure: Performance vs. cost,
security vs. cost.

Comparison
Performance
Elasticity
Cost
Fairness
Scalability
LAN/WAN
Timeliness
Reliability
Availability
throughput
SMICloud A
Framework for
Comparing and
Ranking Cloud
Services
CloudCmp
Comparing Public
Cloud Providers
To Move or Not to
Move The economics
Cloud computing
Performance and
Cost Assessment
of Cloud Services.
Performance
Analysis of
Cloud Computing

Discussion
 Selection factors of performance: Client may select a cloud service
provider according to
 Accountability, Agility(rate of change metrics).
 assurance of service.
 Cost, Performance.
 Security, Privacy.
 Usability.
 Cost factor of public cloud: Every cloud providers has its own cost
model. Most important factor of cloud costing are.
 Initia cost.
 Lease period.
 QoS.
 Age of resource.
 Maintenance cost.
 Data security risk.

Open Issues
 performance: Many service provider are acting. But there is no unique rules
and prerequisition for them. For this reason they have lack of proper services on
 Security (Authentication, Authorization, Access control).
 Privacy.
 Workload.
 Processor power.
 Latency (waiting time).
 Scalability.
 Buffer overflow (Request reject).
 Storage.
 Availability.
 Cost: Due to lack of standartd for costing model a service provier gain more
revenue. But they may not provide proper service according to SLAs. Other
hand clients are not getting proper service but paying more comparing the
service.

Summery
 In my work I have reviewed key concept cost and performance of public cloud.
 The cost model is not feasible for clients but provider.
 Costing model should be regarding to end user.
 Provider should be aware to give more importacy to clients in all aspects
(security, cost etc).
 Provider need to be more concern to their QoS.
 Provider need to reduce the risk of cloud computing which is harmful for user.
 Client need to choose the provider according to over view of provider service.
 In the event of provide outage, it is necessary to ensure data availability
 There need to have a unique cost and performance metrics for public cloud.

Question & answer

24
Thanks to all
For handsome attention!

References:
*http://dazeinfo.com/
**Commerce.gov
*** http://www.computerworlduk.com/it-vendors/microsoft-azure-vs-amazon-aws-public-cloud-
comparison-which-cloud-is-best-for-enterprise-3624848/
1. J. Varia, Best practices in architecting cloud applications in the AWS cloud, in: Cloud Computing: Principles
and Paradigms, Wiley Press, 2011, pp. 459–490. (Chapter 18).
2. State of the Cloud, May 2010. http://www.jackofallclouds.com/2010/05/state-of-thecloud-may-2010
3. Microsoft Windows Azure. http://www.microsoft.com/windowsazure.
4. Google AppEngine. http://code.google.com/appengine.
5. Amazon EC2, http://aws.amazon.com/ec2/
6. S. Dutta, M. Zbaracki and M. Bergen, “Pricing Process as a Capability: A Resource-Based Perspective”,
Strategic Management Journal, vol. 27, no. 7, (2003).
7. S. Maxwell, “The Price is Wrong: Understanding What Makes a Price Seem Fair and the True Cost of Unfair
Pricing”, Wiley, (2008).
8. B. Sharma, R. K. Thulasiram, P. Thulasiraman, S. K. Garg and R. Buyya, “Pricing Cloud Compute
Commodities: A Novel Financial Economic Model”, Proc. of IEEE/ACM Int. Symp. on Cluster, Cloud and Grid
Computing, (2012).
9. L. Xia, K. B. Monroe and J. L. Cox, “The Price Is Unfair! A Conceptual Framework of Price Fairness
Perceptions”, J. of Marketing, vol. 68, (2004).
10. E. Iveroth, A. Westelius, C. Petri, N. Olve, M. Coster and F. Nilsson, “How to Differentiate by Price:
Proposal for a Five-Dimensional Model”, European Management Journal, (2012).

References
11. International Journal of Grid and Distributed Computing Vol.6, No.5 (2013),
http://dx.doi.org/10.14257/ijgdc.2013.6.5.09
12. Amazon Web Services, http://aws.amazon.com/.
13. G. Moore, “Cramming More Components onto Integrated Circuits”, Electronics, vol. 38, no. 8, (1965).
14. W. Wang, P. Zhang, T. Lan and V. Aggarwal, “Datacenter Net Profit Optimization with Individual Job Deadlines”,
Proc. Conference on Inform. Sciences and Systems, (2012).
15. M. Macias and J. Guitart, “A Genetic Model for Pricing in Cloud Computing Markets”, Proc. 26th Symp. of Applied
Computing, (2011).
16. M. Mihailescu and Y. M. Teo, “Dynamic Resource Pricing on Federated Clouds”, Proc. 10th IEEE/ACM Int. Symp.
on Cluster. Cloud and Grid Computing, (2010).
17. A framework for ranking of cloud computing services, journal homepage: www.elsevier.com/locate/fgcs
.
.
21. Cloud metering and billing IBM
22. Cost breakdown of Public Cloud Computing and Pricing Strategy for Cloud Computing Services
23. CloudCmp Comparing Public Cloud Providers
24. Performance Analysis of Cloud Computing
25. Performance and Cost Assessment of Cloud Services
26. SMICloud A Framework for Comparing and Ranking Cloud Services
27.To Move or Not to Move The Economics of Cloud Computing
28. E. Iveroth, A. Westelius, C. Petri, N. Olve, M. Coster and F. Nilsson, “How to Differentiate by Price:
Proposal for a Five-Dimensional Model”, European Management Journal, (2012).

Performance and Cost Analysis of Modern Public Cloud Services

More Related Content

What's hot (20)

Similar to Performance and Cost Analysis of Modern Public Cloud Services (20)

Performance and Cost Analysis of Modern Public Cloud Services

Editor's Notes