CSE 461: Cloud Computing
Lecture 1
Introduction
Prof. Mamun, CSE, HSTU
Our story…
2
Our Data Now…
Personal
Data Emails, Calendars, Contacts,
Documents and Media Location Information, etc…
IoT 3
We Live in a World of Data…
4
The World of Data
5
The World of Data
Internet of Things (IoT)
6
Big Data
• Big data is defined as large pools of data that can be captured,
communicated, aggregated, stored, and analyzed.
• Data continues to grow:
In mid-2010, the information universe carried 2 zettabytes and
2025 predictions expect nearly 90 times more at 181 zettabytes
coming our way.
• Applications are becoming data-intensive.
7
What Do We Do With Data?
Store Share
Access Process
…. and
Encrypt more!
We want to do these seamlessly...
8
Using Diverse Interfaces & Devices
Desktops
Mobile Devices
…and even appliances
Consumer Electronics
We also want to access, share and process our data
from all of our devices, anytime, anywhere!
9
What About the Future?
How will you…
…work on documents? …get your news & info?
…create, access, store
and share media? …navigate?
…communicate with …live in an intelligent
friends and family? home?
…
10
11
…How Will We Manage Our Data?
Manage it
• Personal, but time consuming.
ourselves?
How would you get • Would you keep it on your devices?
access to your data
wherever you are? • or would you keep it online?
What if it’s managed • and you can get this “service” for
by someone else? free or with a subscription?
12
Has this Happened Before?
Innovation Product Service
13
Think of it this Way …
Evolution of water Utility
Get a continuous
Generate your own Buy it as a product and supply of the utility
utility manage it through a dedicated
connection
14
How About Electricity?
Transformation from a Product to a Service
Innovation Product Service
New Disruptive Buy and Maintain Electric Grid, pay only
for the electricity you
Technology the Technology use
15
…and Banking?
Evolution of Banking
Traditional Banking
No Banks
Banking Instruments Internet Banking
(Take care of your
(Give your money (Cheques / Credit (…more services)
own money )
to the bank) Cards)
16
So What is Cloud
Computing?
17
Can We Define Cloud Computing?
“Cloud Computing is the transformation of
IT from a product to a service”
Innovation Product Service
18
Cloud Computing
Transformation of IT from a Product to a Service
Innovation of IT IT Products Cloud Computing
New Disruptive Buy and Maintain On-Demand IT
services on a Pay-as
Technology the Technology You-Go basis
19
So… how would you transform
information technology into a
Service?
20
Requirements to Transform IT
to a Service
Connectivity Ease of Programmability
For moving data around Ease of development of
Interactivity complex services to users
Seamless interfaces Manage Large Amounts
Reliability of Data
Big Data
Failure will affect many
people, not just one Efficiency
Performance Cost
It should not be slower or Power
less efficient than what Scalability & Elasticity
people already have Flexible and rapid
Pay-as-you-Go response to changing user
Should not pay an upfront needs
fee for the service
21
Requirements to Transform IT
to a Service
22
Combine the Enabling
Technologies…
23
Cloud Computing
Think of it as Internet Computing
Computation done over the Internet
• High Bandwidth
• High Speed Internet
Enabling • Virtualization Cloud
Technologies • Utility Computing
Computing
• …
24
… for a more complete
definition!
Cloud Computing is the
delivery of computing as a
service rather than a
product,
whereby shared
resources, software, and
information are provided to
computers and other
devices,
as a metered service over
a network.
25
Why Cloud Computing?
Pay-as-You-Go Simplified IT Scale quickly Flexible options Resource Carbon
economic model management and effortlessly • Configure software Utilization is Footprint
• Reduce capital • All you need is • Resources can be packages, instance improved decreased
expenditure access to the rented and released types operating
systems. • Reduce Idle • Sharing of resources
• No upfront cost internet. as required resources by sharing means less servers,
• Reduced Time to • It’s the providers • Software Controlled • Any software platform
and conolidation less power and less
Market responsibility to • Instant scalablility • Access from any emissions.
• Better utilization of
manage the details. machine connected
CPU / Storage and
to the Internet
Bandwidth.
26
Applications Enabled by Cloud
Computing
High Growth Applications Aperiodic Bursting Applications
Startup Seasonal
Businesses Businesses
On-Off Applications Periodic Applications
Changing
Research computational
Computing patterns over
time
27
High Growth Applications
2001 2006
vs.
Could not keep up with the growth of Growing exponentially
their number of users.
What do you do when your startup
gains attraction?
Can you grow quick
enough?
28
High Growth Applications
Animoto’s Facebook Plugin doubled traffic to the
site every 12 hours for 3 days.
They could scale from 50 servers to 3500 and go
back down using cloud computing services
Users use it to
produce video
pieces from their What do you do when your startup
photos, video clips gains traction?
and music.
Can you grow quick
enough?
29
Aperiodic Bursting Applications
Website went down on 9/11/2001 due to traffic
February 14th – Busiest Day of the Year
US Holiday Season
Website crashed within 10 minutes of the free trouser
promotion during Superbowl 2010
Even if you design your website infrastructure to handle
peak loads, wont it be idle during other times?
30
On-Off Applications
Researchers running large-scale scientific
simulation using 1000s of computers.
Modern Drug Discovery
• Data-intensive simulation and tests to
discover new compounds
• Large compute power required for
simulation jobs
• Time to market is crucial
Why not rent computer time to run
these simulations?
31
Periodic Applications
Sock Market Analysis
• Different compuational requirements
over time
• Mine market data during the day.
• Process and Analyze at night.
Dynamic and Flexible infrastructure can
reduce costs and improve performance.
32
Technical Challenges
Programming is tricky but improving
Tools are continuously evolving
Moving large data is still expensive
Security
Quality of Service
Green computing
Internet Dependence
33
Non-Technical Challenges
Vendor Lock-In
Non-standardized
Security Risks
Privacy
Legal
Service Level Agreements
34
Reference Books
• Distributed and cloud computing:
from parallel processing to the
internet of things
Kai Hwang, Jack Dongarra and
Geoffrey C. Fox, 2013
35
Questions?
36
