You are on page 1of 34

1

Cloud Computing and


Aneka Platform
Cloud Computing
and Aneka Platform
Dr. Rajkumar Buyya
Cloud Computing and Distributed Systems (CLOUDS) Lab
Dept. of Computer Science and Software Engineering
The University of Nelbourne, Australia
www.cloudbus.org
www.buyya.com
www.manjrasoft.com
Najor Sponsors/Supporters
Manjrasoft
3
The Next Revolution in IT
The Big Switch in IT
Classical Computing
Buy & Own
Hardware, System
SoItware, Applications
oIten to meet peak needs.
nstall, Configure,
Test, Verify, Evaluate
Manage
..
Finally, use it
$$$$....$(High CapEx)
Cloud Computing
$ubscribe
Use
$ - pay for what you
use, based on Qo$

;
e
r
y

1
8

m
o
n
t
h
s
?
4
utline
Computer Utilities"
Vision and Promising T Paradigms/Platforms
Cloud Computing and Related Paradigms
Trends, Definition, Cloud Benefits and Challenges
NarketOriented Cloud Architecture
$-oriented Resource llocation
Global Cloud Exchange
Aneka: Cloud Application Platform
Summary and Thoughts for Future
S
omputer Utilities Vision:
Implications of the Internet
1363 - Leonard Kleinrock, ARPANET project
s of now, computer networks are still in their
infancy, but as they grow up and become
sophisticated, we will probably see the spread
of Computer Utilities, which, like present
electric and telephone utilities, will service
individual homes and offices across the
country.
During the last 40 years, several advances
have taken place in both computing and
communications areas that are turning the
vision of Computer Utilities in to a reality.
6
omputing Paradigms and
Attributes: Realizing the `omputer
Utilities Vision
Web
Data Centres
Utility Computing
Service Computing
Crid Computing
P2P Computing
NarketOriented
Computing
Cloud Computing
.
Ubiquitous
Reliable
Scalable
Autonomic
namic
discover
omposable
"oS
SLA

Paradigms
Attributes/Capabilities
Trillion $ business
7
utline
Computer Utilities"
Vision and Promising T Paradigms/Platforms
Cloud Computing and Related Paradigms
Trends, Definition, Cloud Benefits and Challenges
NarketOriented Cloud Architecture
$-oriented Resource llocation
Global Cloud Exchange
Aneka: Cloud Application Platform
Summary and Thoughts for Future
8
Too popular: too man are
In Search of loud omputing
egend:
rid computing
CIoud computing
3
efining louds: There are man
views for what is cloud computing?
Over 20 definitions:
http://cloudcomputing.sys-con.com/read/612375_p.htm
Renting remote storage backup
Renting remote server hosting Web server
Renting remote more servers to manage large workload
Buyya's Scientific definition of Cloud Computing
Cloud is a market-oriented distributed computing system
consisting of a collection of inter-connected and virtuaIised
computers that are dynamicaIIy provisioned and presented
as one or more unified computing resources based on service-
IeveI agreements (SLA) established through negotiation
between the service provider and consumers.
$ = {negotiated and agreed Qo$ parameters +
rewards + penalties for violation of agreement....}
10
$ubscription-Oriented Cloud $er;ices:
compute, apps, data, ..}
as a Service (..aaS)
Clients
Other
Cloud $er;ices
Go;t.
Cloud $er;ices
Pri;ate
Cloud
Cloud
Manager
Public Cloud
11
loud Services
!nfrastructure as a Service
(!aaS)
CPU, $torage: mazon.com,
Nirvanix, GoGrid..
Platform as a Service (PaaS)
Google pp Engine, Microsoft
zure, Manjrasoft neka..
Software as a Service (SaaS)
$alesForce.Com
Infrastructure as a Service (IaaS)
Software as a Service (SaaS)
Platform as a Service (PaaS)
Manjrasoft
....
12
loud eploment Models
Pri;ate/nterprise
Clouds
Cloud computing
model run
within a company's
own Data Center /
infrastructure for
internal and/or
partners use.
Public/Internet
Clouds
3rd party,
multi-tenant Cloud
infrastructure
& ser;ices:
* a;ailable on
subscription basis
(pay as you go)
ybrid/Mixed CIouds
Mixed usage of
pri;ate and public
Clouds:
Leasing public
cloud ser;ices
when pri;ate cloud
capacity is
insufficient
13
loud Applications
WScientific/Tech Applications
Wusiness Applications
WConsumer/$ocial Applications
$.ien.e and Te.ni.al Appli.ations
Business Appli.ations
Consumer/$o.ial Appli.ations
14
hen will loud spending become
50% of IT spending or reach to a
trillion $ business/ear?
120?
2016
15%
600?
2020?
30%
1000?
2020?
50%
uyya's Guestimate!
2000
1S
loud omputing hallenges:
ealing with too man issues
&m, I am not quite
cIear.Yet anoter
compIex IT paradigm?
BiIIing
&tiIity & Risk
Management
ScaIabiIity
ReIiabiIity
Software Eng.
CompIexity
Programming Env.
& AppIication Dev.
16
utline
Computer Utilities"
Vision and Promising T Paradigms/Platforms
Cloud Computing and Related Paradigms
Trends, Definition, Cloud Benefits and Challenges
NarketOriented Cloud Architecture
$-oriented Resource llocation
Global Cloud Exchange
Aneka: Cloud Application Platform
Summary and Thoughts for Future
17
Marketriented loud Architecture:
"oS negotiation and SLAbased Resource Allocation
18
loud to expand (lease more
resources) with increase in demand
13
loud to shrink (unlease resources)
with decrease in demand
20
Man loud fferings: Good, but new issues
vendor lock in, scaling across clouds
CompIex decisions
to make?
Manjrasoft Aneka
IBM CIoud
21
Interloud: ederation of louds
for Scaling Application Services
$torage Cloud
Compute Cloud
$torage Cloud
Compute
Cloud
Directory
Bank
uctioneer
IobaI CIoud
Excange
Enterprise
Resource
Manager
(Proxy)
Broker 1
Enterprise T Consumer
Publish Offers
Request
Capacity
Negotiate/Bid
Broker N
.
.
.
.
22
utline
Computer Utilities"
Vision and Promising T Paradigms/Platforms
Cloud Computing and Related Paradigms
Trends, Definition, Cloud Benefits and Challenges
NarketOriented Cloud Architecture
$-oriented Resource llocation
Global Cloud Exchange
Aneka: Cloud Application Platform
Summary and Thoughts for Future
23
Aneka: The loud Application Platform
(AP) for ResourceIntensive Apps
(Available as Manjrasoft Product)
Lightweight Container hosting
multiple services.
All programming models
available from within the
same container (unlike others)
SDK containing AP!s for
multiple programming
models and tools
%untime Environment for
managing application
execution management
Suitable for
Development of Enterprise
Cloud pplications
Cloud enabling legacy
applications
Portability for Customer Apps:
Enterprise Public Clouds
.NET/Win Mono/inux
Private Cloud
LAN network
mazon
Microsoft Google
BM
Data Center
ardware Profile Services
+ J=EAH
P
e
r
s
i
s
t
e
n
c
e
Task
Model
Thread
Model
Map Reduce
Model
ther
Models
.NET @ indows Mono @ Linux
S
e
c
u
r
i
t

!rogramming Models
Software Development Kit
Management
Studio
Application
Foundation Services
Membership
Services
Reservation
Services
License
Services
A!s
Design
Explorer
Management Kit
Administration
!ortal
SLA-Negotiation
Web Services
Management
Web Services
Storage
Services
Accounting
Services
Fabric Services
namic Resource Provisioning Services
nfrastructure
Phsical Machines/Virtual Machines
Private Cloud
LAN network
Private Cloud
LAN network
mazon
Microsoft Google
BM
Data Center
mazon
Microsoft Google
BM
Data Center
ardware Profile Services
+ J=EAH
P
e
r
s
i
s
t
e
n
c
e
Task
Model
Thread
Model
Map Reduce
Model
ther
Models
.NET @ indows Mono @ Linux
S
e
c
u
r
i
t

!rogramming Models
Software Development Kit
Management
Studio
Application
Foundation Services
Membership
Services
Reservation
Services
License
Services
A!s
Design
Explorer
Management Kit
Administration
!ortal
SLA-Negotiation
Web Services
Management
Web Services
Storage
Services
Accounting
Services
Fabric Services
namic Resource Provisioning Services
nfrastructure
Phsical Machines/Virtual Machines
Patent
(PT)
24
Aneka and our Glass Model
(man in one, one in man)
TP/IP
nternet rchitecture
neka rchitecture
multiple networks
multiple applications
Multi-core Cluster Grid Cloud
Thread Task ... MapReduce


Aneka
multiple applications
2S
Aneka: components
public DumbTask: ITask
{
.
public void Execute()
{
..
}
}
for(int i=0; i<n; i++)
{
.
DumbTask task = new DumbTask();
app.SubmitExecution(task);
}
Executor
$cheduler
Executor
Executor Executor
Client
gent
work units
internet
internet
neka enterprise Cloud
Client
gent
work units
Aneka User Agent
Aneka Worker
Service
Aneka Nanager
Programming / DepIoyment ModeI
26
Aneka Support for Interloud /
ederation of Resource Pools
XenServer Pool
Provisioning over
private Cloud
managed by Xen
$erver
vNWare Pool
Provisioning over
private Cloud
managed by VMWare
Amazon EC2 Pool
Provisioning over
public Cloud provider:
mazon EC2
Executors/Schedulers
Executor
lient Libraries
Executors
Scheduler
Public Cloud
publicl available resources
(phsical and virtual)
Scheduler
internet
Pri;ate
Cloud
private enterprise
network
VPN
(virtual resources)
Executor
Aneka ase Studies
28
(GoFront: A unit of China
$outhern Railway Group)
neka utilizes idle desktops
() to decrease task time
from days to ours
Time
(in rs)
SingIe
Server
Aneka
CIoud
Raw ocomotive Design Files
(Using utoDesk Maya)
Using Maya
Graphical Mode
Directly
Case 1: $ingle $erver
4 cores
server
neka Maya Renderer
&8e private
Aneka Cloud
GoFront Private Aneka
Cloud
LAN network
(Running Maya Batch Mode on demand)
Case 2: neka
Enterprise Cloud Manjrasoft
Application: Locomotive design CAD rendering
23
A scalable architecture for TitanStrike on
line Gaming Portal
%itanStrike Private Aneka
Cloud
LAN network
(Running Game plugin8 on Demand)
Case 2: neka Enterprise Cloud = ScaIabiIity
neka-based
GameController
The local scheduler
interacts with Aneka
and distributes the load
in the cloud.
Manjrasoft
Distributed
log parsing
logs
logs
Case 1: $ingle $erver = uge OverIoad
$ingle scheduler
controlling the
execution of all the
matches.
ame Servers
Gamers profiles
Plaers statistics
Team plaing
Multiple games
Titan $trike On ine
Gaming Portal
Centralized
log parsing
logs
logs
$ingle
GameController
30
ealth are: SaaS loud for
EG Sensor ata Analsis
Lmbedded 8lueLooLh
Lnabled
uaLa rocessor
CommunlcaLlon Module
LCC Sensor Module
LCC uaLa
Analysls
SofLware
Wireless ECG $ensor
$
a
a
$
P
a
a
$

a
a
$
B
l
u
e
t
o
o
t
h

c
o
n
n
e
c
t
i
v
i
t
y
Wireless/ Mobile 3G Network

User
Millions of users
U
s
e
r

R
e
q
u
e
s
t
s
User Requests
&8er QoS ba8ed Scaling of
Compute Re8ource8
Dynamic $calable Runtime
31
IEEE Scalable omputing hallenge
at Grid 2010 onference
32
utline
Computer Utilities"
Vision and Promising T Paradigms/Platforms
Cloud Computing and Related Paradigms
Trends, Definition, Cloud Benefits and Challenges
NarketOriented Cloud Architecture
$-oriented Resource llocation
Global Cloud Exchange
Aneka: Cloud Application Platform
Summary and Thoughts for Future
33
Summar
Several Computing Paradigms have promised to
deliver Computing Utilities" vision:
Cloud Computing is the most recent kid in the block
promising to turn vision into reality
Many exciting business and consumer applications
enabled.
Aneka is rapidly emerging as a platform for:
Building Clouds (especially private) and their apps.
ccelerating pplications and Managing their
execution.
lready used in engineering, life science, gaming,
and business intelligence applications.
34
References
Quick Overview:
Rajkumar Buyya and Karthik $ukumar,
Platforms for Building and Deploying
pplications for Cloud Computing, C$
Communications, Vol.35, No. 1, Pages: 6-11,
$$N 0970-647X, Computer $ociety of ndia
(C$), Mumbai, ndia, May 2011.
Blueprint Paper!
R. Buyya, C. $. Yeo, $. Venugopal, J. Broberg,
. Brandic, CIoud Computing and Emerging
IT PIatforms: Vision, ype, and ReaIity for
DeIivering Computing as te
t
&tiIity,
Future Generation Computer Sy8tem8 (FGC$)
Journal, June 2009.
Aneka:
C. Vecchiola, X. Chu, and R. Buyya, neka:
$oftware Platform for .NET-based Cloud
Computing, igh Speed and Large Scale
Scientific Computing, 267-295pp, O$ Press,
msterdam, Netherlands, 2009.
iley Press, New York, U$A,
March 2011

You might also like