Concepts of Distributed Computing & Cloud Computing

Introduction to Distributed
Computing

Contents
• Introduction
– Motivation
– History
– Goal
– Characteristics
• Example of Applications
• Scenario
• Advantages and Disadvantages
• Issues and Challenges
• Conclusion
• References

Introduction
• Definition
– “A distributed system consists of multiple
autonomous computers that communicate
through a computer network.
– “Distributed computing utilizes a network of many
computers, each accomplishing a portion of an
overall task, to achieve a computational result
much more quickly than with a single computer.”
– “Distributed computing is any computing that
involves multiple computers remote from each
other that each have a role in a computation
problem or information processing.”

Introduction
• A distributed system is one in which hardware or
software components located at networked computers
communicate and coordinate their actions only by
message passing.
• In the term distributed computing, the word
distributed means spread out across space.
Thus, distributed computing is an activity performed on a
spatially distributed system.
• These networked computers may be in the same
room, same campus, same country, or in different
continents

Introduction
Cooperation
Cooperation
Cooperation
Internet
Large-scale
Application
Resource
Management
Subscription
Distribution
Distribution Distribution
Distribution
Agent
Agent Agent
Agent
Job Request

Motivation
• Inherently distributed applications
• Performance/cost
• Resource sharing
• Flexibility and extensibility
• Availability and fault tolerance
• Scalability
• Network connectivity is increasing.
• Combination of cheap processors often more cost-
effective than one expensive fast system.
• Potential increase of reliability.

History
• 1975 – 1985
– Parallel computing was favored in the early years
– Primarily vector-based at first
– Gradually more thread-based parallelism was introduced
– The first distributed computing programs were a pair of
programs called Creeper and Reaper invented in 1970s
– Ethernet that was invented in 1970s.
– ARPANET e-mail was invented in the early 1970s and
probably the earliest example of a large-scale distributed
application.

History
• 1985 -1995
– Massively parallel architectures start rising and message
passing interface and other libraries developed
– Bandwidth was a big problem
– The first Internet-based distributed computing project
was started in 1988 by the DEC System Research Center.
– Distributed.net was a project founded in 1997 -
considered the first to use the internet to distribute data
for calculation and collect the results,

History
• 1995 – Today
– Cluster/grid architecture increasingly dominant
– Special node machines eschewed in favor of COTS
technologies
– Web-wide cluster software
– Google take this to the extreme (thousands of
nodes/cluster)
– SETI@Home started in May 1999 - analyze the radio
signals that were being collected by the Arecibo Radio
Telescope in Puerto Rico.

Goal
• Making Resources Accessible
– Data sharing and device sharing
• Distribution Transparency
– Access, location, migration, relocation, replication,
concurrency, failure
• Communication
– Make human-to-human comm. easier. E.g.. :
electronic mail
• Flexibility
– Spread the work load over the available machines
in the most cost effective way
• To coordinate the use of shared resources
• To solve large computational problem

Characteristics
• Resource Sharing
• Openness
• Concurrency
• Scalability
• Fault Tolerance
• Transparency

Architecture
• Client-server
• 3-tier architecture
• N-tier architecture
• loose coupling, or tight coupling
• Peer-to-peer
• Space based

• Examples of commercial application :
– Database Management System
– Distributed computing using mobile agents
– Local intranet
– Internet (World Wide Web)
– JAVA Remote Method Invocation (RMI)
Application

Distributed Computing Using Mobile Agents
• Mobile agents can be wandering around in a network using free
resources for their own computations.

Local Intranet
• A portion of Internet that is separately administered & supports
internal sharing of resources (file/storage systems and printers) is
called local intranet.

Internet
• The Internet is a global system of interconnected computer networks that use
the standardized Internet Protocol Suite (TCP/IP).

JAVA RMI
• Embedded in language Java:-
– Object variant of remote procedure call
– Adds naming compared with RPC (Remote Procedure Call)
– Restricted to Java environments
RMI Architecture

Categories of Applications in
distributed computing
• Science
• Life Sciences
• Cryptography
• Internet
• Financial
• Mathematics
• Language
• Art
• Puzzles/Games
• Miscellaneous
• Distributed Human Project
• Collaborative Knowledge Bases
• Charity

Example of applications
• Internet – Gomez Distributed PEER Client
(peerReview)
– Evaluate the performance of large websites to find
bottlenecks.
• Life Sciences - Compute Against Cancer® (CAC)
– Create immediate impact in the lives of cancer
patients and their families today, while at the
same time empowering the research that will
result in improved therapies — and perhaps even
the cure.

Example of applications
Collaborative Knowledge Bases – Wikipedia
 A collaborative project to produce a complete a free
encyclopedia from scratch.
 The encyclopedia is available in many non-English languages.
Distributed Human Projects- Open Mind Indoor
Common Sense
 Help teach indoor mobile robots to be smarter. It will
create a repository of knowledge which will enable
people to create more intelligent mobile robots for use
in home and office environments.

• The key characteristics of a variety of modern computing scenarios distinguishing between
micro, medium and global-scale scenarios.
• Micro scale
– Researches focus on small computer-based components that can be deployed in an environment and
can coordinate their actions (i.e. local sensing and effecting of specific environmental conditions)
with the goal of enriching our physical word with specific “smart” functionalities. Potential
applications range from simple monitoring activities, to smart materials and self assembly of
computational beings
• Medium scale
– Our world is populated by local ad-hoc networks (e.g. the ensemble of Bluetooth enabled devices we
could carry on or we could find in our cars) and network-based furniture (e.g., Web enabled fridges
and ovens able to interact with each other and effectively support our cooking activities)
• Global scale
– In Internet computing, there is need to access data and services according to a variety of patterns,
independently of the availability/location of specific servers
Scenario

Scenario
Mechanisms of Self-organization in Modern
Distributed Computing Scenarios

Advantages
• Economics:-
– Computers harnessed together give a better price/performance ratio
than mainframes.
• Speed:-
– A distributed system may have more total computing power than a
mainframe.
• Inherent distribution of applications:-
– Some applications are inherently distributed. E.g., an ATM-banking
application.
• Reliability:-
– If one machine crashes, the system as a whole can still survive if you
have multiple server machines and multiple storage devices
(redundancy).
• Extensibility and Incremental Growth:-
– Possible to gradually scale up (in terms of processing power and
functionality) by adding more sources (both hardware and software).
This can be done without disruption to the rest of the system.

Disadvantages
• Complexity :-
– Lack of experience in designing, and implementing a
distributed system. E.g. which platform (hardware and OS) to
use, which language to use etc.
• Network problem:-
– If the network underlying a distributed system saturates or
goes down, then the distributed system will be effectively
disabled thus negating most of the advantages of the
distributed system.
• Security:-
– Security is a major hazard since easy access to data means
easy access to secret data as well.

Issues and Challenges
• Heterogeneity of components :-
– variety or differences that apply to computer hardware,
network, OS, programming language and
implementations by different developers.
– All differences in representation must be deal with if to do
message exchange.
– Example : different call for exchange message in UNIX
different from Windows.
• Openness:-
– System can be extended and re-implemented in various
ways.
– Cannot be achieved unless the specification and
documentation are made available to software developer.
– The most challenge to designer is to tackle the complexity
of distributed system; design by different people.

Issues and Challenges cont…
• Transparency:-
– Aim : make certain aspects of distribution
are invisible to the application programmer
; focus on design of their particular
application.
– They not concern the locations and details
of how it operate, either replicated or
migrated.
– Failures can be presented to application
programmers in the form of exceptions –
must be handled.

• Transparency:-
– This concept can be summarize as shown in
this Figure:

• Security:-
– Security for information resources in
distributed system have 3 components :
a. Confidentiality : protection against disclosure to
unauthorized individuals.
b. Integrity : protection against alteration/corruption
c. Availability : protection against interference with the
means to access the resources.
– The challenge is to send sensitive
information over Internet in a secure
manner and to identify a remote user or
other agent correctly.

Issues and Challenges cont..
• Scalability :-
– Distributed computing operates at many
different scales, ranging from small Intranet
to Internet.
– A system is scalable if there is significant
increase in the number of resources and
users.
– The challenges is :
a. controlling the cost of physical resources.
b. controlling the performance loss.
c. preventing software resource running out.
d. avoiding performance bottlenecks.

• Failure Handling :-
– Failures in a distributed system are partial –
some components fail while others can function.
– That’s why handling the failures are difficult
a. Detecting failures : to manage the presence of failures cannot
be detected but may be suspected.
b. Masking failures : hiding failure not guaranteed in the worst
case.
• Concurrency :-
– Where applications/services process
concurrency, it will effect a conflict in operations
with one another and produce inconsistence
results.
– Each resource must be designed to be safe in a
concurrent environment.

What is Distributed
Computing/System?
• Distributed computing
–A field of computing
science that studies
distributed system.
–The use of distributed
systems to solve
computational
problems.

• Distributed system
– Wikipedia
• There are several autonomous computational
entities, each of which has its own local
memory.
• The entities communicate with each other by
message passing.
– Operating System Concept
• The processors communicate with one another
through various communication lines, such as
high-speed buses or telephone lines.
• Each processor has its own local memory.

What is Distributed
Computing/System?
• Distributed program
–A computing program that runs in a
distributed system
• Distributed programming
–The process of writing distributed program

What is Distributed Computing/System?
• Common properties
– Fault tolerance
• When one or some nodes fails, the whole
system can still work fine except
performance.
• Need to check the status of each node
– Each node play partial role
• Each computer has only a
limited, incomplete view of the system. Each
computer may know only one part of the
input.

– Resource sharing
• Each user can share the computing power
and storage resource in the system with
other users
– Load Sharing
• Dispatching several tasks to each nodes can
help share loading to the whole system.
– Easy to expand
• It id easy to expand the system by adding
nodes.

Why Distributed Computing?
• The nature of application
• Performance
– Computing intensive
• The task could consume a lot of time on computing.
– Data intensive
• The task that deals with a lot mount or large size of files. For
example, Facebook,Orkut,Twitter ,etc.
• Robustness
– No SPOF (Single Point Of Failure)
– Other nodes can execute the same task executed on
failed node.

Common Architectures
• Communicate and coordinate works among
concurrent processes
– Processes communicate by sending/receiving
messages
– Synchronous/Asynchronous

Common Architectures
• Master/Slave architecture
– Master/slave is a model of
communication where one device
or process has unidirectional
control over one or more other
devices
• Database replication
–Source database can be
treated as a master and the
destination database can
treated as a slave.
• Client-server
–web browsers and web
servers

Best Practice /Characteristics
• Data Intensive or Computing Intensive
– Data size and the amount of data
• The attribute of data you consume
• Computing intensive
–We can move data to the nodes where
we can execute jobs
• Data Intensive
–We can separate/replicate data to
difference nodes, then we can execute
our tasks on these nodes
–Reduce data replication when executing
tasks

• Master nodes need to know data location
• No data loss when incidents happen
–SAN (Storage Area Network)
–Data replication on different nodes
• Synchronization
–When splitting tasks to different nodes, how
can we make sure these tasks are
synchronized?

• Robustness
– Still safe when one or partial nodes fail
– Need to recover when failed nodes are online.
No further or few action is needed
– Failure detection
• When any nodes fails, master nodes can
detect this situation.
–Eg: Heartbeat detection
– App/Users don’t need to know if any partial
failure happens.
• Restart tasks on other nodes for users

 Network issue
 Bandwidth
 Need to think of bandwidth when copying files from one
node to other nodes if we would like to execute the task on
the nodes if no data in these nodes.
 Scalability
 Easy to expand
 Hadoop – configuration modification and start daemon
 Optimization
 What can we do if the performance of some nodes
is not good?
 Monitoring the performance of each node
 According to any information exchange like heartbeat or
log
 Resume the same task on another nodes

summary
• The concept of distributed computing is the most
efficient way to achieve the optimization.
• Distributed computing is anywhere : intranet, Internet
or mobile ubiquitous computing
(laptop, PDAs, pagers, smart watches, hi-fi systems)
• It deals with hardware and software systems, that
contain more than one processing / storage and run in
concurrently.
• Main motivation factor is resource sharing; such as files
, printers, web pages or database records.
• Grid computing and cloud computing are form of
distributed computing.

Case study - Condor
Case study - Hadoop

The History and Future of Cloud
Computing

What is Cloud Computing?
• Cloud Computing is a general term used to describe a new class of
network based computing services that takes place over the
Internet,
– basically a step on from Utility Computing
– a collection/group of integrated and networked
hardware, software and Internet infrastructure (called
a platform).
– Using the Internet for communication and transport
provides hardware, software and networking services
to clients
• These platforms hide the complexity and details of the underlying
infrastructure from users and applications by providing very simple
graphical interface or API (Applications Programming Interface).

What is Cloud Computing?
• In addition, the platform provides on demand
services, that are always
ON, Anywhere, Anytime and Any place.
• Pay for use and as per needed, elastic
– scale up and down in capacity and functionalities
• The hardware and software services are
available to
– general public, enterprises, corporations and
businesses markets

Cloud Summary
• Cloud computing is an umbrella term used to refer to
Internet based development and services
• A number of characteristics define cloud
data, applications services and infrastructure:
– Remotely hosted: Services or data are hosted on
remote infrastructure.
– Ubiquitous: Services or data are available from
anywhere.
– Commodified: The result is a utility computing model
similar to traditional that of traditional utilities, like
gas and electricity - you pay for what you would want!

What is Cloud Computing
Adopted from: Effectively and Securely Using the Cloud Computing Paradigm by peter Mell, Tim Grance
• Shared pool of configurable computing resources
• On-demand network access
• Provisioned by the Service Provider

Cloud Computing Characteristics
Common Characteristics:
Low Cost Software
Virtualization Service Orientation
Advanced Security
Homogeneity
Massive Scale Elastic Computing
Geographic Distribution
Essential Characteristics:
Resource Pooling
Broad Network Access Rapid Elasticity
Measured Service
On Demand Self-Service

Essential Characteristics of cloud computing
• • On-demand self-service: A client can provision
computer resources without the need for interaction with
cloud service provider personnel.
• • Broad network access: Access to resources in the cloud
is available over the network using standard methods in a
manner that provides platform-independent access to
clients of all types.
• This includes a mixture of heterogeneous operating
systems, and thick and thin platforms such as
laptops, mobile phones,.
• • Resource pooling: A cloud service provider creates
resources that are pooled together in a system that
supports multi-tenant usage.

Benefits of cloud computing
• • Rapid elasticity: Resources can be rapidly and
elastically provisioned.
• Measured service: The use of cloud system
resources is measured, audited, and reported to the
customer based on a metered system.

Cloud Service Models
Software as a
Service (SaaS)
Platform as a
Service (PaaS)
Infrastructure as a
Service (IaaS)
Google App
Engine
SalesForce CRM
LotusLive

Type of Cloud
• Deployment models: This refers to the location
and management of the cloud's infrastructure.
• Service models: This consists of the particular
types of services that you can access on a cloud
computing platform.

Classification of Clouds
• Cloud services can also be categorized into
three types based on access and location
– A public cloud is available to anyone on the
Internet
– Any user can sign up to use the public cloud
(e.g. Microsoft Windows Azure)
– A private cloud is a proprietary cloud
environment that only provides cloud
services to a limited number of users
– Private clouds are usually within your own
data center behind your firewall
Slide 59

Classification of Clouds
Slide 60
– A hybrid cloud, sometimes called a
virtual private cloud, provides
services that run on a public cloud
infrastructure, but limits access to it
with a virtual private network (VPN)
[Oracle 2009]

Public Cloud
• The main benefits of using a public cloud service are:
– Managed Infrastructure Easy and inexpensive set-up because
hardware (+storage) is managed by the provider
– Scalability/Elasticity  you get more/less processing power, storage,
etc. when needed
– Eliminates complex procurement cycles improving the time-to-
market for its users
– Removes Undifferentiated "Heavy Lifting“  let its users focus on
delivering differentiating business value instead of wasting valuable
resources on the undifferentiated heavy lifting that makes up most of
IT infrastructure
– Pay as you go  No wasted resources because you pay for what you
use
• Public clouds are run by third parties, and applications from
different customers are likely to be mixed together on the cloud’s
servers, storage systems, and networks.

Public Cloud
• The main drawbacks of using a public cloud
service are
– Vendor lock-in for PaaS  e.g. you cannot easily
change your software environment (e.g. from
Windows Azure with C# to Google App Engine
with Java)
• Examples of public clouds
– Amazon Elastic Compute Cloud (EC2)
– Google App Engine
– Microsoft Windows Azure

Private Cloud
• The main benefits of using a private cloud service are
– Exclusive use providing the utmost control over
data, security, and quality of service
– Ownership  has control over infrastructure and how
applications are deployed
– Managed by companies own IT organization
– High level of control
• The main drawbacks are
– Infrastructure costs (e.g. capital expenses for storage)
– Operating costs for running the data center
– Average utilization is not always known in advance  right
scale?
• Examples of private clouds
– Eucalyptus

Hybrid Cloud
• The main benefits of using a hybrid cloud
service are
– Augment a private cloud with the resources of a
public cloud  provide on-demand, externally
provisioned scale [Oracle 2009]
• The main drawbacks are
– Complexity how to distribute applications
across both a public and private cloud [Oracle
2009]
Slide 64
[Oracle 2009]

Concepts of Distributed Computing & Cloud Computing

Advantage of cloud computing
• • Lower costs: Because cloud networks operate at higher
efficiencies and with greater utilization, significant cost
reductions are often encountered.
• • Ease of utilization: Depending upon the type of service being
offered, you may find that you do not require hardware or
software licenses to implement your service.
• • Quality of Service: The Quality of Service (QoS) is something
that you can obtain under contract from your vendor.
• • Reliability: The scale of cloud computing networks and their
ability to provide load balancing and failover makes them highly
reliable, often much more reliable than what you can achieve in
a single organization.

• • Outsourced IT management: A cloud computing
deployment lets someone else manage your computing
infrastructure while you manage your business. In most
instances, you achieve considerable reductions in IT
staffing costs.
• • Simplified maintenance and upgrade: Because the
system is centralized, you can easily apply patches and
upgrades. This means your users always have access to
the latest software versions.
• • Low Barrier to Entry: In particular, upfront capital
expenditures are dramatically reduced. In cloud
computing, anyone can be a giant at any time.
Advantage of cloud computing

Disadvantages of Cloud Computing
• Requires a constant Internet connection:
– Cloud computing is impossible if you cannot
connect to the Internet.
– Since you use the Internet to connect to both your
applications and documents, if you do not have an
Internet connection you cannot access anything,
even your own documents.
– A dead Internet connection means no work and in
areas where Internet connections are few or
inherently unreliable, this could be a deal-breaker.

• Does not work well with low-speed connections:
– Similarly, a low-speed Internet connection, such as
that found with dial-up services, makes cloud
computing painful at best and often impossible.
– Web-based applications require a lot of bandwidth to
download, as do large documents.
• Features might be limited:
– This situation is bound to change, but today many
web-based applications simply are not as full-featured
as their desktop-based applications.
• For example, you can do a lot more with Microsoft
PowerPoint than with Google Presentation's web-based
offering

• Can be slow:
– Even with a fast connection, web-based applications
can sometimes be slower than accessing a similar
software program on your desktop PC.
– Everything about the program, from the interface to
the current document, has to be sent back and forth
from your computer to the computers in the cloud.
– If the cloud servers happen to be backed up at that
moment, or if the Internet is having a slow day, you
would not get the instantaneous access you might
expect from desktop applications.

• Stored data might not be secure:
– With cloud computing, all your data is stored on the
cloud.
• The questions is How secure is the cloud?
– Can unauthorised users gain access to your confidential
data?
• Stored data can be lost:
– Theoretically, data stored in the cloud is safe, replicated
across multiple machines.
– But on the off chance that your data goes missing, you
have no physical or local backup.
• Put simply, relying on the cloud puts you at risk if the cloud lets
you down.

Measuring the Cloud's Value
• • Profit: The economies of scale can make this a profitable business.
• • Optimization: The infrastructure already exists and isn't fully utilized.
• This was certainly the case for Amazon Web Services.
• • Strategic: A cloud computing platform extends the company's products
and their franchise.
• This is the case for Microsoft's Windows Azure Platform.
• • Extension: A branded cloud computing platform can extend customer
relationships by offering additional service options.
• This is the case with IBM Global Services and the various IBM cloud
services.

• • Presence: Establish a presence in a market before a large
competitor can emerge.
• Google App Engine allows a developer to scale an
application immediately. For Google, its office applications
can be rolled out quickly and to large audiences.
• • Platform: A cloud computing provider can become a hub
master at the centre of many ISV's (Independent Software
Vendor) offerings.
Measuring the Cloud's Value

Basic Cloud Characteristics
• The “no-need-to-know” in terms of the underlying details of
infrastructure, applications interface with the infrastructure
via the APIs.
• The “flexibility and elasticity” allows these systems to scale
up and down at will
– utilising the resources of all kinds
• CPU, storage, server capacity, load balancing, and databases
• The “pay as much as used and needed” type of utility
computing and the “always on, anywhere and any place”
type of network-based computing.

Basic Cloud Characteristics
• Cloud are transparent to users and applications, they can be
built in multiple ways
– branded products, proprietary open
source, hardware or software
– In general, they are built on clusters of PC servers
and off-the-shelf components plus Open Source
software combined with in-house applications
and/or system software.

Accessing the Cloud
• Cloud APIs allow software to request data and
computations from one or more services through a direct
or indirect interface.
• Cloud APIs most commonly expose their features via REST
and/or SOAP. (Representational state transfer)
• Vendor specific and cross-platform interfaces are available
for specific functions.
• Cross-platform interfaces have the advantage of allowing
applications to access services from multiple providers
without rewriting, but may have less functionality or other
limitations vs vendor-specific solutions

Cloud API Available for
• Infrastructure
• Infrastructure APIs modify the resources available to operate the
application.
• Service
• Service APIs provide an interface into a specific capability provided by a service
explicitly created to enable that capability.
• Database, messaging, web portals, mapping, e-commerce and storage are all
examples of service APIs.
• Application
• Application APIs provide methods to interface and extend applications on
the web.
• Application APIs connect to applications such as CRM, ERP, social media
and help desk.

Dropbox
• Free space: 2GB
• Premium space: $99/year for
100GB
• File size limit: Unlimited
• Platforms:
Windows, Mac, Linux, iOS, Andr
oid, BlackBerry
• Best for: Seamless syncing
Google Drive
 Free space: 15GB
 Premium space: $59.88/year for
100GB
 File size limit: 10GB
 Platforms: Windows, Mac, iOS,
Android
 Best for: Web apps
Apple iCloud
 Free space: 5GB
 Premium space: $100/year for 50GB
 File size limit: 25MB free/250MB
paid
 Platforms: Mac, iOS, Windows
 Best for: Heavy iTunes/Mac users
Microsoft SkyDrive
 Free space: 7GB
 Premium space: $50/year for 100GB
 File size limit: 2GB
 Platforms:
Windows, Mac, iOS, Android, Windows
Phone
 Best for: Windows/Office integration
Free space

• All 3 – Dropbox, Google Drive and SkyDrive – start with a
certain amount of free space allotted to every user, which
can be increased upon payment, with rates varying for
each provider.
• SkyDrive starts with the maximum amount of free space –
7GB. Google starts with 15GB, while Dropbox starts with
the least – 2GB. You can however, considerably increase
your Dropbox storage capacity – by referrals, beta
testing, camera upload through your phone, and many
other tweaks. Or if you happen to buy an HTC device, you
start off with 25GB of space straightaway.
• If you’re not planning to pay for your storage – SkyDrive or
Dropbox offer the most value.

File Type Support
• Any file type can be uploaded on to these cloud services – but you can only view
file types that are supported. Keeping that in mind, here’s a comparison on file
type support for the three platforms:
• Dropbox doesn’t support any file type. All files must be downloaded and nothing
can be opened online. It’s not a major issue though – if you’re using Dropbox on
your phone – you can edit files right from your phone (via an editor of course) and
have them updated. The same would apply for your computer.
• Google Drive supports unusual, and in a way, diverse range of file types – like
Autodesk AutoCAD files, Photoshop (.psd) files, and even Adobe Illustrator files.
But at the same time, it lacks basics. You can only view, but not edit Microsoft
Office documents. All such files are converted to their Google Docs equivalent for
editing. That can troublesome if you’re using a phone. Since Google Docs can only
be accessed online, there is hardly any offline usability. Serious disadvantage here.
• SkyDrive, being Microsoft’s child, will let you open and edit any Office documents.
There is even support for audio formats, but limited to MP4 and WMV only.

Online features• Dropbox is the oldest of the three providers, and is thus more
established. It is bound to attract more users thanks to features like
referral, which earn you additional space.
• It is also the only service to support Linux and Blackberry systems.
• It even has passcode locks for its mobile apps.
• Although it charges higher dollar-per-GB, it offers a clean interface,
is easier to use and integrates very nicely with your phone.
• Purchasing an HTC gets you 25GB of free space. If you own an HTC,
a Blackberry or run a Linux – Dropbox is the place to go.

Google Drive
• Google Drive offers support more file types which makes online
sharing and editing much easier.
• It also offers more value for dollar-per-GB.
• Plus, if you purchase additional space, you get 25GB extra on to
your Gmail storage.
• But the fact that you have to convert office documents to Google
Docs before you can begin editing can be a big pain.

SkyDrive
• SkyDrive gives you the most free space to start with. It also lets
you remote access your PC online.
• SkyDrive belonging to Microsoft, it has a head start in terms of
targeting Windows users.
• SkyDrive integrates superbly with Windows Phone and Windows
8.
• Office 2013’s default file-saving location is your SkyDrive account.
• It may be termed an unfair advantage, but if you have a Windows
Phone or have recently purchased Windows 8, SkyDrive is a pretty
good option.

Apple iCloud
• iCloud isn't much of a Dropbox competitor, but if all you need synced between
your devices are text documents, it can be a pretty seamless solution.
• Various apps such as Pages and iA Writer have iCloud sync capabilities, saving
your work after every keystroke and instantly sending changes to Apple's servers.
• Once you open up iA Writer or Pages on another Apple device or Mac with OS X
Lion, you'll already be working with the most recent version of your document.
• Additionally, Lion saves versions of your documents locally using Time Machine
so you can return to older versions of your document, but only on your machine.
• While iCloud is a very elementary document-syncing solution, it might also be
the simplest one to use. And if you need to stream music or videos you've
purchased from the cloud, you can do that, too

Enterprise Applications
• Characteristics of Enterprise Applications
–complex (business) logic,
–huge, complex data,
–special security requirements,
–focus is on transactions ,
–many clients,
–support for different platforms,
–heterogeneous development environments:
C, C++, Java, C#, COBOL, ...
• Application Requirements are a driving force for
Software Architecture.
Slide 102

Mainframe Applications
• Approach
– Mainframe provides all
resources and logic
– Terminals for user
interaction (thin clients)
• Advantages
– Simple deployment
• Problems and Limitations
– Limited GUI capabilities
– Every client allocates
server resources 
– Limited scalability
Slide 103
Mainframe:
Application and Data
Terminals: UI

Client/Server Architecture• Approach
– Increased capabilities of PCs
– Logic was (partially)
transferred to clients  thick
clients
– Central database server
• Advantages
– Better user experience (GUIs)
• Problems and Limitations
– Every client has a permanent
stateful connection
– Every client holds server
resources (db connections, …)
– Limited scalability
Slide 104
Server: Data
Thick Clients : UI + Logic
PC
Mac Linux-
Workstation

Middle-Tier Architecture
• Approach
– Logic bundled in Middle-tier
– Clients have no permanent
connection to Middle-tier 
stateless connections
• Advantages
– Factoring of business logic
– Middleware provides services
(pooling, security, …)
– Resources are shared between
clients  improved scalability
• Problems
– New programming model
Slide 105
Thin Clients
Data-Tier
Middle-Tier:
Logic + Services
(Middleware)

Stateful Services
• The service host provides a service instance for
every client during the whole lifetime of a session.
Slide 106
• Comparable to client/server architecture
– The services is an “extension” of the client
• Technologies
– EJB: Stateful Session Beans
– WCF: Per-Session Services
– CORBA
– Java RMI
Client 1
Proxy
Service Host
Instance 1
Instance 2
Client 2
Proxy

Stateful Services – Problems
• Every service instance holds server resources
– Only a limited number of clients can be served.
• What happens with the service instance when the client
dies unexpectedly?
– Distributed garbage collection, session timeouts, …
• What was the state of the instance when the connection
was interrupted?
– Reestablishing a connection may be painful.
• Transactions may manage state differently
– Transactions should be bound to service operations, not to
sessions.
– When the transaction is rolled back, database state and
session state are not in synch.
Slide 107

Stateless Services
• A service instance is created for each invocation of a
service operation.
Slide 108
Client 1
Proxy
Service Host
Instance 1
Instance 2
Instance 3
Client 2
Proxy
 There is no permanent connection between the client and
the service
 Resources can be shared between clients 
 Improved scalability.
 Technologies
 EJB: Stateless Session Beans
 WCF: Per-Call Services

Internet-Scale Application
• 2011 stats:
– +200B pageviews/month
– >3.9T feed actions/day
– +300M active users
– >1B chat mesgs/day
– 100M search queries/day
– >6B minutes spent/day (ranked #2
on Internet)
– +20B photos, +2B/month growth
– 600,000 photos served / sec
– 25TB log data / day processed
– 120M queries /sec
• Scaling the “relational” data:
– Keeps data
normalized, randomly
distributed, accessed at high
volumes
– Uses “shared nothing”
architecture

Internet-Scale Application
• 2011 stats:
– +20 petabytes of data processed / day by +100K MapReduce jobs
– 1 petabyte sort took ~6 hours on ~4K servers replicated onto ~48K disks
– +200B clusters, each at 1-5K nodes, handling +5 petabytes of storage
• ~40 GB/sec aggregate
read/write throughput
across the cluster
• +500 servers for each search
query < 500ms
• Scaling the process:
– MapReduce: parallel
processing framework
– BigTable: structured hash
database
– Google File System: massively
scalable distributed storage

Microsoft in the Cloud
(15 years)
450M+
active users
(13 years)
550M
users/mth
(12 years)
Largest non-
ICP/IP cloud
service
x100M users
(11 years)
320M+
active
users
(11 years)
2B
queries/mth
(15 years)
450M+
active users
(7 years)
5B conf
min/yr
(6 years)
4B emails/day

Windows Azure Platform
• The Windows Azure Platform is a group of
cloud technologies to be used by applications
running in Microsoft’s data centers, on-
premises and on various devices.
• Windows Azure is a platform for running
Windows applications and storing data in the
cloud.

Windows Azure applications run in Microsoft
data centres and are accessed via the Internet.

Working
• Windows Azure runs on machines in Microsoft data
centres.
• Rather than Microsoft customers can install and run
software on their own computers , Windows Azure provide
a service.
•
• By which customers use it to run applications and store
data on Internet-accessible machines owned by Microsoft.
• Those applications might provide services to businesses, to
consumers, or both.

Applications that might be built on Windows
Azure
• An independent software vendor (ISV) :
-It could create an application that targets business users, an approach
that’s often referred to as Software as a Service (SaaS).
- ISVs can use Windows Azure as a foundation for a variety of
business-oriented SaaS applications.
- An ISV might create a SaaS application that targets consumers.
- -Facebook, OLX ,etc.
• Enterprises might use Windows Azure to build and run applications
that are used by their own employees.

Three main components of Windows Azure :

Windows Azure has three main components:
• Compute : The Compute service runs applications .
• Storage : The Storage service stores data .
• Fabric: It provides a common way to manage and monitor
applications that use this cloud platform.

Physical Machine 1
Compute Resources: Web and Worker Roles
• Web Role: Interactive .NET application hosted in IIS:
– Web Application or Web Service (WCF)
• Worker Role: Durable background process
– Often isolated from outside world
– There are ways to make it reachable for external application
• Fabric Agent collects resource metrics (usage, failures, …)
Slide 120
Hardware
Load
Balancer
Virtual Machine 2
Fabric Controller
Virtual Machine 1
Web Role
Instance
IIS
Fabric Agent
Physical Machine 2
Virtual Machine 2
Virtual Machine 1
Worker Instance
Fabric Agent
HTTP/HTTPS

Web Role
• Web role instance can accept incoming HTTP
or HTTPS requests.
• To allow this, it runs in a VM that includes
Internet Information Services (IIS) 7.
• Developers can create Web role instances
using ASP.NET, WCF, or another .NET
technology that works with IIS.

Worker Role
• It can’t accept requests from the outside world.
• A Worker role instance initiates its own requests for
input.
• It can read messages from a queue and it can open
connections with the outside world.
• Worker role instances can be viewed as similar to a
batch job or a Windows service.

Instance1
Instance 2
Instance 3
Instance 4
Instance 5

Windows Azure Storage provides
Blobs, Tables, and Queues.
Blob: binary large object

Window Azure Storage
• Cloud Applications can store data in
– Blobs
– Tables
– Queues
• Blobs provide storage for unstructured data.
– Blobs are organized in containers.
• Tables are structured collections of entities.
– Entities are collections of name/value pairs.
– Tables have no schema.
• Queues provide storage for small data items on a FIFO
basis.
– Queues enable communication between service instances.
Slide 125

Windows Azure Storage Architecture
Slide 126
Windows Azure Storage
Cloud Application/External Application
Storage Account
Blob Container
REST API
Blob Blob
Queue Table
Entity
Entity
Entity
.NET API (StorageClient)
Blob Blob
REST API

Storage Features
• Windows Azure Storage
– provides huge storage resources,
– is massively scalable, and
– is a highly reliable persistence mechanism.
• Data is stored in large server.
• Scalability
– Data can be distributed across many storage nodes.
– Storage access is load-balanced.
• Reliability
– Data is replicated to different storage nodes when a write
operation occurs (3 replicas in different default domains).
– When a device fails, data is replicated to a new storage node.
Slide 127

Storage Account
• The storage account is the entry point for all storage
services.
• Storage accounts can be created on the Azure
Development Portal.
• Storage accounts can be accessed with secret key:
– On the client side a hash code is computed (with SHA256).
–The hash code and the secret key is used to compute a
HMAC-SHA256 signature (Hash Message Authentication
Code).
– The HMAC is attached to the REST request.
– The storage server uses the signature to authenticate the
request.
Slide 128

Accessing Storage via REST
• Every storage item is identified by a Request URI:
– Blob: <http|https>://<account name>.blob.core.windows.net/<container>/<blob name>
– Table: <http|https>://<account name>.table.core.windows.net/<table name>
– Queue: <http|https>://<account name>.queue.core.windows.net/<queue name>
• HTTP verb represents the action executed on the resource:
– PUT: create container/queue/table, save blob/entity/queue item, set metadata, …
– GET: list blobs in container, get blob/entity/queue item, …
– DELETE: delete container/queue/table/blob/entity/queue item, …
• The REST API provides a platform-independent interface to
Azure storage.
• Example:
Slide 129
PUT http://myaccount.blob.core.windows.net/mycontainer/myblockblob HTTP/1.1
x-ms-version: 2009-09-19
x-ms-date: Sun, 27 Sep 2009 22:33:35 GMT
Content-Type: application/octet-stream
x-ms-blob-type: BlockBlob
x-ms-meta-m1: some metadata
Authorization: SharedKey myaccount:YjN4fAR8/AmBrqBz7MG2uFinQh4dscbj598g=
Content-Length: nn
Request Body

Blobs
• Blob Containers provide a grouping of a set of
blobs.
– Sharing policies are set at container level: public
read or private.
– E.g. Facebook Album(Public , Friends, Private)
– Containers can have metadata.
– E.g. Date /Time of creation
• Blob store big chunks of data.
– Blobs can have metadata.
Slide 130
Storage Account
Blob Container 1
Page
Blob 1
Page
Blob 2
Block Blob 1
B1 B2 B3
Block Blob 1
B1 B2 B3
Blob Container 2

Block Blob
blocks
Page Blob
pages
Twotypes ofBlobs

Page Blob – Random Read/Write

Blob cont…….
• Windows Azure Blob storage is a service for
storing large amounts of unstructured data that
can be accessed from anywhere in the world via
HTTP or HTTPS.
• Common uses of Blob storage include:
• Serving images , videos or documents directly to a browser
• Storing files for distributed access
• Streaming video and audio
• Performing secure backup and disaster recovery

BlobStorageConcepts
BlobContainerAccount
cohowinery
images
PIC01.JPG
PIC02.JPG
videos VID1.AVI
http://<account>.blob.core.windows.net/<container>/<blobname>

Blob Containers
Multiple Containers per Account
Special $root container
Blob Container
A container holds a set of blobs
Set access policies at the container level
Associate Metadata with Container
List the blobs in a container
IncludingBlobMetadata andMD5
NOsearch/query.i.e.noWHEREMetadataValue = ?
Blobs Throughput
Effectively in Partition of 1
Target of 60MB/s per Blob

BlobFeatures andFunctions
PutBlob
GetBlob
DeleteBlob
CopyBlob
<name, value>

TableStorageConcepts
EntityTableAccount
cohowinery
Comments
Name =…
Comment= …
Advertisements
URI =…
Description =…
Name =…
Comment= …
URI =…
Description =…

Table . . .Table Table
Windows Azure Storage
A closer look at tables
Entity . . .Entity Entity
Property PropertyProperty
Name Type Value
Storage
Accounts

Tables
• Tables provide structured storage.
• Tables contain a set of entities.
• Entities contain a set of properties.
• Properties have
– a name, and
– a typed value (Int32, Int64, double, string, bool,
DateTime, byte array).
• Tables have no fixed schema, i.e. the structure of
each entity may be different.
• Each entity has two key properties:
– The partition key, and
– the row key.
Slide 143
Storage Account
Table1
Table 2
Entity 1
Property 1
Property 2
…
Entity 2
Property 1
Property 2
…

Partitioning of Tables
• The application controls the distribution of
entities by assigning partition keys to entities.
• If two entities have different partition keys they
may be stored in different storage nodes.
• Only entities stored within the same table and the
same partition can take part in an entity group
transaction.
• Exp: Partition key 2= GHRCE
- All friends Name and DOB is
Store in Node 2.
If Partition key 1= Shivaji College
- Then all friends Name and
DOB is Store in Node 1.
Slide 144
Storage Node 2
Person
Name Meyers
DOB 1999/5/5
Row Key 1
Partition Key 2
Name Gates
Row Key 2
Partition Key 2
Storage Node 1
Person
Name Gosling
DOB 1955/9/6
Row Key 3
Partition Key 1
DOB 1955/4/3

Entity Properties
Entity can have up to 255 properties
Up to 1MB per entity
Mandatory Properties for every entity
PartitionKey & RowKey
Uniquely identifiesan entity
Definesthe sortorder
No fixed schema for other properties
Each property is stored as a <name, typed value> pair
No schema stored for a table

TableOperations
Create, Delete
Insert
Update
Merge
Replace
Delete

QueueStorageConcepts
MessageQueueAccount
order processing
customer ID
order ID
http://…
customer ID
order ID
http://…
cohowinery

Queue
Using Queues
Web Role
ASP.NET,
WCF, etc.
Worker Role
main()
{ … }
1) Receive
work
2) Put work
in queue
3) Get work
from queue
4) Do
work

Queues
• Queues provide a reliable asynchronous message
delivery mechanism.
• Reliability
– Messages are redelivered until a consumer
confirms that it has been successfully processed
– Messages are processed at least once.
• Decoupling of producer and consumer
– Different parts of the system can be implemented
using different technologies and languages.
– Producer and consumer needn’t be alive at the
same time.
• Scalability
– Queue length grows  number of consumer
instances can be increased.
Slide 154
Message
Producer
Message
Consumer
CloudQueue
Message
CloudQueue
Message
Queue

Inter-role Communication
• Role instances can communicate asynchronously via queues.
– Preferred method for reliable messaging
Slide 155
Role
Instance 1
Queue
Role
Instance 2
Role
Instance 3
Role
Instance 4
 Role instances can also communicate directly using TCP or HTTP(S)
connections.
External
App
Role
Instance 2
Load
Balancer
Role
Instance 1
Role
Instance 3
Role
Instance 4
External
Endpoint
Internal
Endpoint

Queues – Features and Limitations
• Queue
– The number of messages is not limited.
– Metadata can be associated.
– There is no guaranteed return order of messages.
– Messages are delivered at least once  consumer must be able to
deal with messages that are delivered more than once.
• Message
– Message can be up to 8KB in size.
– Message structure:
• MessageID: a GUID
• VisibilityTimeout: If processing of message is not confirmed within this time
range the message reappears in the queue (default = 30 seconds)
• MessageTTL: Time-to-live interval for messages (maximum = default = 7
days)
• Payload: String or byte array.
Slide 156

LooselyCoupledinteraction withQueues
Azure Queue
Input Queue
(Work Items)
Web
Role
Web
Role
Web
Role

• All Windows Azure applications and all of the
data in Windows Azure Storage live in some
Microsoft data center.
• Within that data center, the set of machines
dedicated to Windows Azure is organized into
a fabric(structure, framework ).
• Windows Azure Fabric consists of a (large)
group of machines, all of which are managed
by software called the fabric controller.

Cont…
• The fabric controller is replicated across a
group of five to seven machines.
• Because it can communicate with a fabric
agent on every computer, it’s also aware of
every Windows Azure application in this fabric.
• It monitors all running applications, for
example, giving it an up-to-the-minute picture
of what’s happening in the fabric.

• To create a scalable Web application on Windows Azure, a developer can use Web roles and tables.
CREATING A SCALABLE WEB APPLICATION

• The clients are browsers, and so the application logic
might be implemented using ASP.NET or another Web
technology.
• It’s also possible to create a scalable Web application
that exposes RESTful and/or SOAP-based Web services
using WCF.
• The fabric controller also monitors these
instances, making sure that the requested number is
always available.
• For data storage, the application uses Windows Azure
Storage tables, which provide scale-out storage capable
of handling very large amounts of data

So what do you mean by Scalable Application?
• In a conventional data center, it requires always having
enough machines on hand to handle the peaks hours
time, even though most of those systems go unused
most of the time.
• But ,if the application is built on Windows Azure, the
organization running it can expand the number of
instances it’s using only when needed, then shrink back
to a smaller number.
• An online IPL ticketing selling site, Customer Call
Centers Video sites with occasional hot
stories, governments online exam web sites that are
used mostly at certain times of day, and others.

A PARALLEL PROCESSING APPLICATION

• The parallel work is done by some number of Worker role
instances running simultaneously, each using blob data.
• Since Windows Azure imposes no limit on how long an
instance can run, each one can perform an arbitrary
amount of work.

• Through this interface, the user might
determine how many Worker instances should
run, start and stop those instances, get
results, and more.
• Communication between the Web role
instance and the Worker role instances relies
on Windows Azure Storage queues.

A parallel processing application can
communicate with an on-premises application.

• Rather than relying on a Web role instance running on
Windows Azure, the user might instead interact with the
Worker role instances via an on-premises application .
• Multiple Worker role instances run simultaneously, each
interacting with the outside world via queues.
• Here, however, work is put into those queues directly by
an on-premises application
• In a scenario like this, the user might have no idea that
the on-premises application he’s using relies on
Windows Azure for parallel processing.
• Exp: mobile tracking application
On-premises software is installed
and run on computers on the
premises (in the building) of the
person or organisation using the
software, rather than at a remote
facility, such as at a server or
cloud somewhere on the
Internet.

CREATING A SCALABLE WEB APPLICATION WITH
BACKGROUND PROCESSING

• A majority of today’s applications built to provide a
browser interface.
• That do nothing but accept and respond to browser
requests that are useful, but they’re also limiting.
• There are lots of situations where Web-accessible
software also needs to initiate work that runs in the
background, independently from the request/response
part of the application.
• For example, Online recommendation systems
, Facebook etc.

• Instead, the part of the application that accepts
browser requests should be able to initiate a
background task that carries out this work
without the knowledge of users.
• Windows Azure Web roles and Worker roles can
be used together to address this scenario. i.e.
Background processing

• The two major parts of Windows Azure are:
– Azure AppFabric
– SQL Azure

AppFabric
• Windows Azure AppFabric provides a comprehensive cloud
middleware platform for developing, deploying and managing
applications on the Windows Azure Platform.
• It enables bridging your existing applications to the cloud
through secure connectivity across network and geographic
boundaries.
• Let existing WhatsApp application merge with Facebook’s cloud
application.
• It delivers additional developer productivity, adding in higher-
level Platform-as-a-Service (PaaS) capabilities on top of the
familiar Windows Azure application model.
Video

• There are three key components of Windows Azure
AppFabric:
• Middleware Services: platform capabilities as
services, which raise the level of abstraction and
reduce complexity of cloud development.
• Composite Applications: a set of new innovative
frameworks, tools and composition engine to easily
assemble, deploy, and manage a composite
application as a single logical entity.
• Scale-out application infrastructure: optimized for
cloud-scale services and mid-tier components.
AppFabric

• Middleware Services
• Windows Azure AppFabric provides pre-built, higher level
middleware services that raise the level of abstraction and
reduce complexity of cloud development.
• These services are open and interoperable across languages
(.NET, Java, Ruby, PHP…) and give developers a powerful
pre-built “class library" for next-gen cloud applications.
• Developers can use each of the services stand-alone, or
combine services to provide a composite solution.
AppFabric

• Service Bus: provides secure messaging and connectivity
capabilities that enable building distributed and disconnected
applications in the cloud, as well as hybrid applications across
both on-premise and the cloud.
• It enables using various communication and messaging
protocols and patterns, and removes the need for the
developer to worry about delivery assurance, reliable
messaging and scale.

• Access Control :enables an easy way to
provide identity and access control to web
applications and services, while integrating
with standards-based identity providers,
including enterprise directories such as
Active Directory®, and web identities such as
Windows Live ID, Google, Yahoo! and
Facebook.

• Caching : performance of Windows Azure and SQL Azure
based apps by providing a distributed, in-memory application
cache, provided entirely as a service (no installation or
management of instances, dynamically increase/decrease
cache size as needed).
• Pre-integration with ASP.NET enables easy acceleration of
web applications without having to modify application code.

• Integration: on Windows Azure, using out-of-box
integration patterns to accelerate and simplify development.
• It will also deliver higher level business user enablement
capabilities such as Business Activity Monitoring and Rules, as
well as self-service trading partner community portal and
provisioning of business-to-business pipelines.

• Composite App : it delivers a complete hosting
environment for web services built using Windows
Communication Foundation (including WCF Data Services and
WCF RIA Services) and workflows built using Windows
Workflow Foundation.
More Info

SQL Azure
• SQL Server hosted in the cloud.
• It provides relational database features, so that a
platform is scalable, highly available and load-balanced.
Video

Concepts of Distributed Computing & Cloud Computing

Recommended

More Related Content

What's hot (20)

Similar to Concepts of Distributed Computing & Cloud Computing (20)

More from Hitesh Kumar Markam (19)

Recently uploaded (20)

Concepts of Distributed Computing & Cloud Computing

Editor's Notes