This document discusses cloud computing and network traffic. It covers several topics: 1. Web-scale problems that are data-intensive and require large data centers to solve. Examples include searching the web and scientific research data. 2. Large data centers have become necessary to handle web-scale problems by centralizing computing resources. Issues include redundancy, efficiency, and management. 3. Different cloud computing models have emerged like utility computing, platform as a service, and software as a service. For all the network traffic generated, smart network access control, traffic management, and effective security techniques are needed, and there is ongoing research in these areas.

1. Cloud Computing & NW Traffic
Dr.S.Sridhar, Ph.D.(JNUD),
RACI(Paris, NICE), RMR(USA),
RZFM(Germany)
Dean – Cognitive & Central Computing
Facility
R.V. COLLEGE OF ENGINEERING
BANGALORE

3. The iSchool
University of Maryland
What is Cloud Computing?
1. Web-scale problems
2. Large data centers
3. Different models of computing
4. Highly-interactive Web applications

4. The iSchool
University of Maryland
1. Web-Scale Problems
 Characteristics:
 Definitely data-intensive
 May also be processing intensive
 Examples:
 Crawling, indexing, searching, mining the Web
 “Post-genomics” life sciences research
 Other scientific data (physics, astronomers, etc.)
 Sensor networks
 Web 2.0 applications
 …

5. The iSchool
University of Maryland
How much data traffic?
 Wayback Machine has 20 TB/month (2006)
 Google processes 60TB /Week (2015)
640K ought to be
enough for anybody.

8. The iSchool
University of Maryland
There’s nothing like more data!
s/inspiration/data/g;

9. The iSchool
University of Maryland
What to do with more data?
 Answering factoid questions
 Pattern matching on the Web
 Works amazingly well
 Learning relations
 Start with seed instances
 Search for patterns on the Web
 Using patterns to find more instances
Who shot Abraham Lincoln? → X shot Abraham Lincoln
Birthday-of(Mozart, 1756)
Birthday-of(Einstein, 1879)
Wolfgang Amadeus Mozart (1756 - 1791)
Einstein was born in 1879
PERSON (DATE –
PERSON was born in DATE

10. The iSchool
University of Maryland
2. Large Data Centers
 Web-scale problems? Throw more machines at it!
 Clear trend: centralization of computing resources in large
data centers
 Necessary ingredients: fiber, juice, and space
 What do Oregon, Iceland, and abandoned mines have in
common?
 Important Issues:
 Redundancy
 Efficiency
 Utilization
 Management

13. The iSchool
University of Maryland
Key Technology: Virtualization
Hardware
Operating System
App App App
Traditional Stack
Hardware
OS
App App App
Hypervisor
OS OS
Virtualized Stack

14. The iSchool
University of Maryland
3. Different Computing Models
 Utility computing
 Why buy machines when you can rent cycles?
 Examples: Amazon’s EC2, GoGrid, AppNexus
 Platform as a Service (PaaS)
 Give me nice API and take care of the implementation
 Example: Google App Engine
 Software as a Service (SaaS)
 Just run it for me!
 Example: Gmail
“Why do it yourself if you can pay someone to do it for you?”

15. The iSchool
University of Maryland
4. Web Applications
 A mistake on top of a hack built on sand held together by
duct tape?
 What is the nature of software applications?
 From the desktop to the browser
 SaaS == Web-based applications
 Examples: Google Maps, Facebook
 How do we deliver highly-interactive Web-based
applications?
 AJAX (asynchronous JavaScript and XML)
 For better, or for worse…

21. The iSchool
University of Maryland
Web-Scale Problems?
 Don’t hold your breath:
 Biocomputing
 Nanocomputing
 Quantum computing
 …
 It all boils down to…
 Divide-and-conquer
 Throwing more hardware at the problem
Simple to understand… a lifetime to master…

22. The iSchool
University of Maryland
Divide and Conquer
“Work”
w1 w2 w3
r1 r2 r3
“Result”
“worker” “worker” “worker”
Partition
Combine

23. The iSchool
University of Maryland
Different Workers
 Different threads in the same core
 Different cores in the same CPU
 Different CPUs in a multi-processor system
 Different machines in a distributed system

24. The iSchool
University of Maryland
Choices, Choices, Choices
 Commodity vs. “exotic” hardware
 Number of machines vs. processor vs. cores
 Bandwidth of memory vs. disk vs. network
 Different programming models

25. The iSchool
University of Maryland
Flynn’s Taxonomy
Instructions
Single (SI) Multiple (MI)
Data
Multiple(M
SISD
Single-threaded
process
MISD
Pipeline
architecture
SIMD
Vector Processing
MIMD
Multi-threaded
Programming
Single(SD)

26. The iSchool
University of Maryland
SISD
D D D D D D D
Processor
Instructions

27. The iSchool
University of Maryland
SIMD
D0
Processor
Instructions
D0D0 D0 D0 D0
D1
D2
D3
D4
…
Dn
D1
D2
D3
D4
…
Dn
D1
D2
D3
D4
…
Dn
D1
D2
D3
D4
…
Dn
D1
D2
D3
D4
…
Dn
D1
D2
D3
D4
…
Dn
D1
D2
D3
D4
…
Dn
D0

28. The iSchool
University of Maryland
MIMD
D D D D D D D
Processor
Instructions
D D D D D D D
Processor
Instructions

29. The iSchool
University of Maryland
Memory Typology: Shared
Memory
Processor
Processor Processor
Processor

30. The iSchool
University of Maryland
Memory Typology: Distributed
MemoryProcessor MemoryProcessor
MemoryProcessor MemoryProcessor
Network

31. The iSchool
University of Maryland
Memory Typology: Hybrid
Memory
Processor
Network
Processor
Memory
Processor
Processor
Memory
Processor
Processor
Memory
Processor
Processor

32. The iSchool
University of Maryland
Parallelization Problems
 How do we assign work units to workers?
 What if we have more work units than workers?
 What if workers need to share partial results?
 How do we aggregate partial results?
 How do we know all the workers have finished?
 What if workers die?
What is the common theme of all of these problems?

33. The iSchool
University of Maryland
General Theme?
 Parallelization problems arise from:
 Communication between workers
 Access to shared resources (e.g., data)
 Thus, we need a synchronization system!
 This is tricky:
 Finding bugs is hard
 Solving bugs is even harder

34. The iSchool
University of Maryland
Managing Multiple Workers
 Difficult because
 (Often) don’t know the order in which workers run
 (Often) don’t know where the workers are running
 (Often) don’t know when workers interrupt each other
 Thus, we need:
 Semaphores (lock, unlock)
 Conditional variables (wait, notify, broadcast)
 Barriers
 Still, lots of problems:
 Deadlock, livelock, race conditions, ...
 Moral of the story: be careful!
 Even trickier if the workers are on different machines

35. The iSchool
University of Maryland
Patterns for Parallelism
 Parallel computing has been around for decades
 Here are some “design patterns” …

36. The iSchool
University of Maryland
Master/Slaves
slaves
master

37. The iSchool
University of Maryland
Producer/Consumer Flow
CP
P
P
C
C
CP
P
P
C
C

38. The iSchool
University of Maryland
Work Queues
CP
P
P
C
C
shared queue
W W W W W

39. The iSchool
University of Maryland
Conclusions
For all these traffic ,
Network access has to be smart
NW traffic has to be controlled
Effective security is needed
A lot of research scope is there in the domains of NW
access, security algorithms
This area is everlasting !