1. Lecture 0. Introduction Instructor: Weidong Shi (Larry), PhD
COSC6376 Cloud Computing
Lecture 0. Introduction
Instructor: Weidong Shi (Larry), PhD
Computer Science Department
University of Houston

2. Topics
Scope of the course
Grading policy
Overview of cloud and computing

3. Web Sites
Class website:

4. Scope of This Course Understand the basic ideas of cloud computing
Get familiar with
Tools
Systems
Expose to some research topics
Complete a team project

5. Prerequisites Linux OS Some programming skills
Java, python, ruby, shell scripting
Comfortable with learning new programming frameworks
Sufficient knowledge about
Data structure and databases
Operating systems
Distributed systems

6. Tentative Schedule Parallel data processing Cloud infrastructures
Distributed file systems (HDFS)
MapReduce
Cloud based databases
High-level distributed data management
Cloud infrastructures
Virtualization
Amazon AWS
Microsoft Azure
Interactive front-end – Google App Engine
Research topics
Resource provisioning
Privacy and security

7. Assignments Reading papers Some simple programming assignments
Individual readings will be posted on the wiki
Need to submit a short summary for some posted papers
Libraries of papers
Some simple programming assignments
Help you master the concepts
Learn to use tools and systems
A team project
Middleware, applications, implementations, theory

8. Course Grading 35% assignments 65% project
15% assigned reading summaries (due before class)
20% two programming assignments
65% project
15% for report one
15% for report two
20% for final report
15% for final presentation

9. Course Project Teams of 2-3 people 3 milestones
We encourage 3 people
3 milestones
1/3: Present a problem statement and literature survey
2/3:Report progress, challenges, …
Final
Paper
In class presentations: last weeks of the semester
1/3 and 2/3 presentations should be prepared

10. What is Cloud Computing?

11. Overview What is meant by Cloud Computing Utility Computing
X as a Service
Infrastructure as a Service
Platform as a Service
Software as a Service
Why do corporations need to pay attention
Applications

12. What is Cloud Computing?
Old idea: Software as a Service (SaaS)
Def: delivering applications over the Internet
Recently: “[Hardware, Infrastrucuture, Platform] as a service”
Utility Computing: pay-as-you-use computing
Illusion of infinite resources
No up-front cost
Fine-grained billing (e.g. hourly)

13. Why need Cloud Computing?
Traditional licensed software
Software as service
With fixed money, you buy all the bells and whistles with using it or not often.
Pay cash up front.
Like leasing a car, you use it but can not make any significant changes.
Pay according to the distance you traveled.

14. Hardware Views NO NEED! Effect foundry on hardware
Cloud computing on companies
Only companies like Intel and Samsung can own fabrication lines.
Foundries enable “fab-less” semiconductor chip companies.
Large companies amortize operational costs.
Similarly, datacenter providers offer service for datacenter-less companies.
NO NEED!

15. Why need Cloud Computing?
Public Cloud: available in a pay-as-you-go way. e.g. Amazon Web Services, Google Engine, and Microsoft Azure.
Private: not available to the public such as internal datacenters of a business or other organization.
Advantages
Service providers: simplified software installation and centralized control.
End users: access the services and share data easily. Store data safely.
Application providers: same as foundries to chip companies.

16. Cloud Computing vs. Grid Computing
Cloud computing = virtualization+ grid + services + utility computing
Grid computing: resource provisioning, load balancing, parallel processing
Views of different users
System admin/hadoop users: grid
Application owners/service users: service, utility

17. Google Trends

18. Google Trends

19. Gartner’s 2011 Hype Cycle

20. Gartner’s 2011 Hype Cycle

22. Driving Forces Behind Cloud
Experience with very large datacenters – profitable for cloud providers
Economics of scale
Pervasive broadband Internet
Fast x86 virtualization
Pay-as-you-go billing model
Large user base
Online payment
Online Ads
Content distribution
 Web 2.0 lowers the entry point to e-business  more small e-business owners
 Large user base of clouds

23. Economics of Scale
How many servers does Google has?

24. Google Server Count

25. Who Owns The Most Servers

26. Perils of Corporate Computing
Own information systems 
However
Capital investment 
Heavy fixed costs 
Redundant expenditures 
High energy cost, low CPU utilization 
Dealing with unreliable hardware 
High-levels of overcapacity (Technology and Labor) 
NOT SUSTAINABLE

28. Google: CPU Utilization
Activity profile of a sample of 5,000 Google Servers over a period of 6 months

29. Google Server Farms (Oregon)

31. Google: Energy Overhead

32. Reading Assignment

33. Cloud Characteristics
On-demand self-service
Ubiquitous network access
Location independent resource pooling
Rapid elasticity
Pay per use

34. Delivery Models Software as a Service (SaaS)
Use provider’s applications over a network
SalesForce.com
Platform as a Service (PaaS)
Deploy customer-created applications to a cloud
AppEng
Infrastructure as a Service (IaaS)
Rent processing, storage, network capacity, and other fundamental computing resources
EC2, S3

35. Software Stack
Mobile (Android), Thin client (Zonbu) Thick client (Google Chrome)
Clients
Identity, Integration Payments, Mapping, Search, Video Games, Chat
Services
Application
Peer-to-peer (Bittorrent), Web app (twitter), SaaS (Google Apps, SAP)
Platform
Java Google Web Toolkit, Django, Ruby on Rails, .NET
Storage
S3, Nirvanix, Rackspace Cloud Files, Savvis,
Infrastructure
Full virtualization (GoGrid), Management (RightScale), Compute (EC2), Platform (Force.com)

36. Cloud Killer Apps Mobile and web applications
Parallel batch processing / MapReduce
Data analytics
OLAP, data mining, machine learning
Extensions of desktop software
Matlab, Mathematica

37. Big Data Every Where! Lots of data is being collected and warehoused
Web data, e-commerce
purchases at department/ grocery stores
Bank/Credit Card transactions
Social Network
The rise of analytics
understanding customers, supply chains, buying habits, ranking, and so on
Cloud based big data analytics
Computation produces small data output containing a high density of information
Implemented in Clouds

38. How Much Data? Google processes 20 PB a day (2008)
Twitter generates approximately 12 TB of data per day
New York Stock Exchange 1TB of data everyday
eBay processes 50 petabytes of data a day

39. Facebook
2.5 billion content items shared per day (status updates + wall posts + photos + videos + comments)
2.7 billion Likes per day
300 million photos uploaded per day
500+terabytes of new data ingested into the databases every day

40. Topics Covered

41. Topics Covered Economics of cloud computing
Tools to create your own cloud infrastructure
Public cloud
AWS, Google Engine
Big data analytics using cloud
Large scale services using cloud
Resource management
Security and privacy

42. Companies Are Afraid to Use Clouds
[Chow09ccsw]

43. Privacy