1. Dr. Elise de Doncker CS6260 Yazeed K. Almarshoud
Cloud Computing
Dr. Elise de Doncker
CS6260
Yazeed K. Almarshoud

2. Roadmap Flynn’s Taxonomy Introduction Parallel vs. Distributed
Grid computing structure
Flynn’s Taxonomy
Cloud vs. Grid
Cloud Computing
Possibilities
Some Characteristics of Cloud Computing
SaaS and Cloud Computing
Supercomputing & Cloud Computing
Clouds Examples
Conclusions
References

3. Introduction
During the good economic times, enterprises do huge investment in Information Technology (IT) infrastructure to achieve faster and reliable response to users’ queries.
The concept of parallel computing & distributing systems widely used and enhanced in many related environments (.i.e Grids)
What is exactly the difference when we say Parallel or Distributed?

4. Parallel vs. Distributed
Parallel computing generally means:
Vector processing of data
Multiple CPUs in a single computer
Distributed computing generally means:
Multiple CPUs across many computers 4

5. Flynn’s Taxonomy SISD Single-threaded process MISD
Instructions
Single (SI)
Multiple (MI)
Single (SD)
SISD
Single-threaded process
MISD
Pipeline architecture
SIMD
Vector Processing
MIMD
Multi-threaded Programming
Data
Multiple (MD) 5

6. SISD
Processor D D D D D D D
Instructions 6

7. SIMD Processor D0 D0 D0 D0 D0 D0 D0 D1 D1 D1 D1 D1 D1 D1 D2 D2 D2 D2… … … … … … … Dn Dn Dn Dn Dn Dn Dn
Instructions 7

8. MIMD Processor D D D D D D D Instructions Processor D D D D D D D8

9. Parallel vs. Distributed
Processor D D D D D D D
Shared
Memory
Instructions
Network connection for data transfer
Processor D D D D D D D
Instructions
Parallel: Multiple CPUs within a shared memory machine
Distributed: Multiple machines with own memory connected over a network 9

10. Divide and Conquer Partition Combine “Work” w1 w2 w3 r1 r2 r3 “Result”
“worker”
“worker”
“worker” r1 r2 r3
Combine
“Result”

11. Grid Computing Structure (big picture)

12. Cloud computing
“Cloud computing is a computing paradigm shift where computing is moved away from personal computers or an individual application server to a “cloud” of computers. Users of the cloud only need to be concerned with the computing service being asked for, as the underlying details of how it is achieved are hidden. This method of distributed computing is done through pooling all computer resources together and being managed by software rather than a human.“

13. Cloud vs. Grid
Cloud Computing is an infrastructure that virtualizes hardware and software resources
Grid Computing are patterns, tools and frameworks to distribute computing or data
A cloud can be the platform to run a computing or data grid

14. Cloud Computing
Cloud computing is a novel platform for computing and storage.
Cloud computing provisions and configures servers as needed.
It allows for more efficient use of the enterprise resources and applications.
It introduces accountability and streamlines computing needs of an enterprise.

15. Possibilities
It is possible to consolidate all the needs of an organization in a systematic and accountable fashion.
It is possible to procure computing related resources similar to how you rent a place for living.
For example,
you can buy storage on demand from amazon.com in a service it offers called the “S3”
You can buy computation service from amazon.com in its “elastic cloud computing” service (EC2)
Usage example: You are in charge of IT in a local company. You have an immediate need for backing up entire set up for a short period of time as a mock up for disaster recovery. What would you do?

16. What is driving Cloud Computing
Technology advances that support massive scalability & accessibility
Emergence of data intensive applications & new types of workloads
Large scale information processing, i.e. parallel computing using Hadoop
Web 2.0 rich media interactions
Light weight run anywhere web apps
Skyrocketing costs of power, space, maintenance, etc.
Explosion of data intensive applications on the Internet
Advances in multi-core
computer architecture
Fast growth of connected mobile devices
Growth of Web 2.0-enabled PCs, TVs, etc.

17. Industry Trends Leading to Cloud Computing
2008
2000
Cloud Computing
Next-Generation Internet computing
Next-Generation Data Centers
1998
Software as a Service
Network-based subscriptions to applications
Gained momentum in 2001
1990
Utility Computing
Offering computing resources as a metered service
Introduced in late 1990s
Grid Computing
Solving large problems with parallel computing
Made mainstream by Globus Alliance

18. Some Characteristics of Cloud Computing
Virtual – Physical location and underlying infrastructure details are transparent to users
Scalable – Able to break complex workloads into pieces to be served across an incrementally expandable infrastructure
Efficient – Services Oriented Architecture for dynamic provisioning of shared compute resources
Flexible – Can serve a variety of workload types – both consumer and commercial

19. Cloud Computing in the New Enterprise Data Center
Software Development
Deploys development tools for immediate use
Technology
Incubation
Reduces time to launch new offerings
Innovation Enablement
Expands sources of innovation, increases competitiveness
Large Scale Information Processing
Optimizes emerging Internet scale workloads
Workload Solution
Patterns
Cloud Computing Management Services
Self-service
Admin Portal
Workload Pattern
Templates
Administration
Workflows
SLA and
Capacity Planning
Workload
Management
Provisioning
Monitoring
Virtualized Physical
Servers
(Ensembles)
iDataPlex, BladeCenter, System x, System p, System z

20. Why Cloud Computing? Pay per use Instant Scalability Security
Reliability
APIs
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

21. Case Study of a Cloud Deployment
Traditional
Cloud
New Development
Liberated funding for new development, transformation investment or direct saving
100%
Software Costs
Strategic
Change
Capacity
Power Costs
Current IT Spend
Deployment (1-time)
Labor Costs (Operations and Maintenance)
Software Costs
The TAP deployment team took an ideal internal environment for a cloud implementation.
Reductions
Hardware, labor and power savings reduced the annual cost of operation by 86.7% ($7.6M)
- Hardware costs were reduced by 88.7% (In pre-cloud environment, 488 new servers were required to support 120 projects. In post-cloud environment, 55 new servers were required to support 120 projects)
- Labor costs were reduced by 80.7% (In pre-cloud environment, 15 admins were required. In post-cloud environment, 2 admins were required)
- Power costs were reduced by 88.8% due to the reduction in number of required servers
Other costs
- Software costs remained relatively flat in the pre-cloud and post-cloud environments (This was a simple virtualization scenario. No application virtualization was involved in the deployment)
- 1-time Deployment costs consisted of software (TPM, ITM, RedHat) and services. (~$657,000)
The reduction in operations costs freed capital to invest in new development, make acquisitions, reduce debt, or pay dividends.
Evaluation Metrics:
Payback period: The initial investment ($657,000 deployment cost) will be recouped in 72 days
NPV: The NPV of $7.5M shows that the discounted inflows over the life of the investment exceed the required investment
IRR: The IRR shows that the rate of return over a 3-year period is 499%
ROI: The ROI shows that this deployment yielded more than an adequate return on invested assets
The numbers here may seem unbelievably good but external research has shown a 1:7 compression ratio with cloud technology. A 1:7 compression means the technology can compress your existing cost down to a 7th, roughly 15% of the pre-cloud cost, consistent with the 80-90% reduction that we see in this case study.
Definitions of a few metrics used in quick investment appraisals:
Savings per year
Payback = cost / cash flow = the time that it takes for the gain to repay the sum of the original investment (generally 2-3 years)
NPV = the magnitude of return on the investment, increase in wealth (greater than 0 is good)
IRR = proxy for the rate of growth or the expected return (greater than the yield of an alternative investment, generally the cost of capital)
ROI = (gain – cost) / cost
Notes:
Negative NPV can be good if it’s less negative than alternatives (e.g. cost center)
IRR undervalues cash flows that occur late in a project's life (e.g. revenue generated by renting spare capacity)
ROI can be easily manipulated, depending on what you include in gain and cost
Transition line : Now that we have the big picture, let’s take a deep dive into their financial statement and see how they did it.
Power Costs
(88.8%)
Hardware Costs (annualized)
Hardware, labor & power savings reduced annual cost of operation by 83.8%
Labor Costs
( %)
Hardware Costs
( %)
Note: 3-Year Depreciation Period with 10% Discount Rate

22. “Cloud Computing” Defined “as a Service” types
Everything as a service (EaaS or XaaS)
Communication as a service (CaaS)
Infrastructure as a service (IaaS)
Monitoring as a service (MaaS)
Software as a service (SaaS – includes Application Service Provider (ASP) services)
Platform as a service (PaaS)
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

23. Infrastructure as a Service
SaaS
Software as a Service
PaaS
Platform as a Service
IaaS
Infrastructure as a Service
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

24. SaaS Software as a Service
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

25. Software delivery model
SaaS
Software delivery model
Increasingly popular with SMEs
No hardware or software to manage
Service delivered through a browser
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

26. Advantages SaaS Pay per use Instant Scalability Security Reliability
APIs
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

27. Examples Commercial Services: SaaS CRM Financial Planning
Human Resources
Word processing
Commercial Services:
Salesforce.com
cloud
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

28. PaaS Platform as a Service
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

29. Platform delivery model
Platforms are built upon Infrastructure, which is expensive
Estimating demand is not a science!
Platform management is not fun!
PaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

30. Popular services PaaS Storage Database Scalability
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

31. Advantages PaaS Pay per use Instant Scalability Security Reliability
APIs
PaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

32. Examples PaaS Google App Engine Mosso AWS: S3
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

33. Infrastructure as a Service
IaaS
Infrastructure as a Service
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

34. Computer infrastructure delivery model
Access to infrastructure stack:
Full OS access
Firewalls
Routers
Load balancing
Sometimes called Utility computing
IaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney] 34

35. Advantages IaaS Pay per use Instant Scalability Security Reliability
APIs
Sometimes called Utility computing
IaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney] 35

36. Examples IaaS Flexiscale AWS: EC2 Sometimes called Utility computing
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney] 36

37. Infrastructure as a Service
SaaS
Software as a Service
PaaS
Platform as a Service
IaaS
Infrastructure as a Service
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

38. Common Factors SaaS PaaS IaaS Pay per use Instant Scalability Security
Reliability
APIs
PaaS
IaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

39. Advantages SaaS PaaS IaaS Lower cost of ownership
Reduce infrastructure management responsibility
Allow for unexpected resource loads
Faster application rollout
PaaS
IaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

40. Cloud Economics SaaS PaaS IaaS Multi-tenented
Virtualisation lowers costs by increasing utilisation
Economies of scale afforded by technology
Automated update policy
PaaS
IaaS
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

41. Supercomputing & Cloud Computing
Two macro strategies dominate large-scale (intentional) computing infrastructures
Supercomputing type Structures
Large-scale integrated coherent systems
Managed for high utilization and efficiency
Emerging cloud type Structures
Large-scale loosely coupled, lightly integrated
Managed for availability, throughput, reliability

42. How should we think about the cloud opportunities?
Virtual zoo of systems?
Replacements for Clusters?
Extensions to existing systems and infrastructure?
Surge capacity?
Edge datasystems?
Opportunity to go “hardwareless” when designing new systems and services?

43. The Virtual Zoo
Access to a diverse image library provides an inexpensive mechanism to test applications and services on a variety of OS configurations without having to build all of them.
Leverages virtualization and community images
Leverages “cloud” when scale is important
Using cloud for scalability testing could be interesting when you have servers you want to stress and test, but limited time and resources
Creating hundreds of running instances is relatively easy and could be done by a few people in less than a day
Automation of the scalability testing could be easily accomplished

44. As Replacements for Clusters?
There have been several experiments creating virtual clusters in EC2 and probably in other environments as well [Peter Skomoroch, et al].
These “soft” clusters are interesting, constructed on demand and then torn down with the application run is complete.
It might be possible to integrate virtual clusters into existing Linux cluster queues such that jobs that are queued for a physical cluster could be dispatched to a local cluster or a cloud based virtual cluster for execution.
In fact for throughput jobs this might be even more effective.
Local facilities that start supporting image based scheduling services would lead in this transition (i.e. you submit your job as one or more images rather than scripts or executables)
Cloud hosting for clusters provides one easy way to implement cycle banking since each application determines their own operation environment and overheads are relatively low
This would ideally be implemented as a distributed resource if physical ownership was important
Virtual ownership would make it much easier and robust to implement

45. Seamless extensions
Like in the previous example seamlessly extending an existing queue could be a one way to integrate clouds with existing services and systems.
But we can imagine others.
How about using the cloud as a giant impedance matcher for geographically distributed systems of large-scale sensors and tightly coupled data analysis environments?
The idea is simple.
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

46. Surge Capacity Power companies have peakers.
Typically natural gas powered turbines used during times of peak demand for power.
Clouds can be used for surge capacity for groups that have variable demands for access to compute cycles or server/service cycles

47. Sensor + Cloud + Supercomputer = Next Generation Simulations
Imagine thousands (or millions) of distributed sensors deployed over the globe each generating data in some asynchronous fashion.
Each sensor updates data structures in the cloud via local internet connections. The cloud is ubiquitous, secure enough, reliable etc. and scales to the size of the sensor network and acts as an impedance matcher.
Periodically harvesting processes (in the cloud say) wake up and organize the datasets into a fashion that they can be downloaded coherently to a supercomputer for data assimilation to a large-scale parallel simulation.

48. Going Hardwareless
Need: 24x7 access to flexibly configured hardware, scalable data infrastructure, and customized operating environment
1000 cores x .10 hour x 8760 hours/year x 3 years = $2.6M
1000 cores x $390/core + 3 x $43,800 power + 3 x 200K + 3 x 100K = $1.4M
In my example if cluster utilization is < 53% then it is cheaper to go “hardwareless” at current retail prices
[An Introduction to SaaS and Cloud Computing presentation By Ross Cooney]

49. Clouds Examples Amazon.com Hadoop (Map/Reduce)
Amazon Simple Storage Service (Amazon S3) .
Amazon Elastic Compute Cloud (Amazon EC2)
Hadoop (Map/Reduce)
Large scale information processing, i.e. parallel computing

50. Conclusions The emerging concept of the cloud is pretty cool.
The existing available “retail” models are hugely empowering, since they require only a credit card to get going.
Ease of use is being tackled, a market is developing for images and value added services.
Clouds feel like the next thing that will have traction and will enable hardwareless ventures.
Scientific applications will not drive clouds, but will benefit from their widespread adoption.
It is a disruptive technology in many ways and the university/agency shift will take some time, hence private sector will likely get significantly ahead.
Many groups should be experimenting and it really is pretty cheap to gain the critical experience to figure out interesting things to try.

51. References http://en.wikipedia.org/wiki/Cloud_computing
Includes references to Amazon, Apple, Dell, Enomalism, Globus, Google, IBM, KnowledgeTreeLive, Nature, New York Times, Zimdesk
Others like Microsoft Windows Live Skydrive important
An Introduction to SaaS and Cloud Computing presentation By Ross Cooney
Policy Issues
Hadoop (MapReduce) and “Data Intensive Computing”
See Data intensive computing minitrack at HICSS-42 January 2009
OGF Thought Leadership blog
OGF22 talks by Charlie Catlett and Irving Wladawsky-Berger

52. Presentation Question:
What are the two macro strategies dominate large-scale (intentional) computing infrastructures? Explain.
Supercomputing type Structures
Large-scale integrated coherent systems
Managed for high utilization and efficiency
Emerging cloud type Structures
Large-scale loosely coupled, lightly integrated
Managed for availability, throughput, reliability