cloud computing basics

UNIT -1
CLOUD COMPUTING BASICS
Prepared by P. Bhavani

Cloud computing
 Cloud computing is the process of delivering or
providing computational resources like software
and/or hardware as a service over the cloud (internet).

 With cloud computing the users can access the
database resources through internet from anywhere in
the world. The databases in the cloud are very
dynamic.

Origin of cloud computing
 This concept of providing resources via global
networks starts from 1960’s.
 This idea was introduced byJ.C.R.Licklider, who was
responsible for ARPANET in 1969.
 After the wide internet usage in 1990’s, large
bandwidth was offered. Thus this concept came to use
by public.
 Contd…….

 first milestones for cloud-computing was the
introduction of salesforce.com in 1999.
 the concept of delivering enterprise applications
through a simple website.
 Amazon Web Services in 2002 was the next major
development in cloud computing, which provided a
group of cloud based services.
 Contd….

 Amazon launched its Elastic Compute Cloud (EC2) in
2002 as a commercial web service - to hire/rent
computers on which they can run their own
applications.
 introduction of web2.0 in 2009. Web2.0 is a website
that allows the users to interact with each other in a
social media creator. Examples of Web2.0 include
social networking websites, blogs etc.

 Google and others started to offer browser based
enterprise applications, through services such as
Google apps.
 virtualization technology, development of universal
high-speed bandwidth and universal software
standards.
 Most of the IT professionals use cloud computing as it
offers increased storage space, high flexibility and very
low cost.

Cloud Components
 There are three cloud components, They are
1. Clients
2. Data centers
3. Distributed servers

Clients
 Clients are the end user devices where the users
interact with cloud to manage their information.
 They are usually computers and also
laptops,notebook computers, tablets, mobile phones,
PDAs etc.

 Clients usually filtered in three categories, they are
1. Mobile clients.
2. Thin clients.
3. Thick clients.

 Mobile clients:This refer to mobile devices including
PDAs or smart phones like iPhone.

 Thin Clients:This refer to computers that do not have
internal hard drives. They allow the server do all the
work, but then display the information on the screen.

 Thick clients:This refers to regular computers, that
uses a web browser like Chrome, Firefox, Internet
Explorer to connect to the internet.

 Data Centers
 The data center is the collection of numerous
servers.
 It could be a large room with full of servers
located anywhere in the world.
 The clients can access these servers through the
cloud.

 Distributed Servers
Servers are in geographically separate locations in the
world.
For the cloud subscriber, these servers act as if they
are very near.
This gives the service provider more flexibility in
security and options.

 For example,Amazon has servers all over the world. If
there is a failure at one site, the service would still be
accessed through another site.

Essential characteristics:
 1) On-Demand self service
 2) Broad network access
 3) Location independent resource pooling
 4) Rapid elasticity
 5) Measured service

On demand self service
 allows user to receive the services such as computing
resources, server time, network storage automatically
without direct interaction with the service provider.
 Some of the popular on demand self service providers
are AWS (Amazon Web
Services), Google, Microsoft, IBM,Salseforce.com.

Broad network access
 This is another essential aspect that is available over
the network. They are accessed by using standard
mechanisms in thick or thin client platforms.

Location independent resource
pooling
 The service providers resources are pooled in order to
serve multiple consumers.
 There is a sense of location independence as the
customer has no control over location where the
resources are provided.
 Consumers need not worry about how the cloud
allocates the provided resources.

Rapid elasticity
 The definition of elasticity is the ability to scale the
resources up and down as required. The storage on
cloud seems to be unlimited for the client. The
consumer can use as much as he needs at any time.

Measured services
 Another essential attribute is that the resources can be
measured, controlled and reported.
 This provides transparency for both provider and
consumer of the used service.
 Metering capability is used to control and optimize
resource use.

Architectural influences
 High Performance Computing
 is the use of super computers and parallel processing
techniques for solving complex computational problems.
 HPC technology focuses on developing parallel processing
algorithms and systems by incorporating both administration
and parallel computational techniques.
 High-performance computing (HPC) evolved due to meet
increasing demands for processing speed.

HPC brings together several technologies such as
 computer architecture,
 algorithms,
 programs and electronics,
 and system software under a single canopy to solve
advanced problems effectively and quickly.

 A highly efficient HPC system requires a high-
bandwidth, low-latency network to connect
multiple nodes and clusters.

 HPC technology is implemented in multidisciplinary
areas including:
Geographical data
Scientific research
Oil and gas industry modelling
Electronic design automation
Climate modelling
Media and entertainment

Utility and enterprise grid
computing
 Utility computing is a service providing model in
which a service provider makes computer resources
and infrastructure management available to the
customer whenever needed.
 The consumer is charged for specific usage of the
resources. It is an On- Demand computing method.

 Renting the computer resources such as hardware,
network bandwidth, software to the customer on
demand basis is called on demand computing.
 This type of approach is also called as Pay per use
model or metered service.

 The method used in utility computing is becoming
popular in enterprise computing. It is sometimes used
for consumer market, internet service, web site access,
file sharing and other applications.

 The main advantage of this model is there is no initial
cost or at-least low cost to get the required computer
resources.
 leaders in utility computing are IBM, HP, Microsoft
and then industries like Google, Amazon etc

Enterprise Grid Computing
 Enterprise-grid computing is a form of computing
particularly inside an enterprise. It is a collection of
network components under the control of a grid
management entity

 This management entity manages the assignment of
resources to services to meet the specific business
goals.
 Enterprise computing is used in single data center or
multiple data centers. Enterprise- grid supports
various types of workloads such as transactions.

 These types of grids are used to meet the specific set of
business goals.
 The services that run on an enterprise-grid may range
from the traditional commercial enterprise
applications such as ERP to new concepts such as
distributed applications.

 Enterprise computing enables the companies to the
following
 To dynamically provide resources.
 To simplify tasks.
 To consolidate computing components.
 To set standards across enterprise.
 To scale the resources and workload.

 Some of the advantages of enterprise-grid computing
are,
 It reduces hardware, software costs
 It also reduces employee costs.
 It improves the quality of service through quick
response time.

Autonomic computing
 Autonomic computing refers to the self managing
characteristics of distributed computing resources,
adapting to unpredictable changes.
 It controls the functioning computer applications and
systems without input from the user. This computing
model has systems that run themselves, capable of
doing high level functions.

 The complexity of the system is invisible to the users.
The concept of autonomic computing was first
introduced by IBM in 2001. This model aims to
develop computer systems capable of self
management. This overcomes the rapidly growing
complexity of computing systems management.

 These systems take decisions on their own by using
high level policies.
 It will constantly check and optimize the status. Thus
it automatically adapts itself to changing conditions.
 An autonomic computing framework is composed of
autonomic components ( AC ) interacting with each
other.

 The following are the main components of autonomic
computing:
 Two main control loops ( local and global )
 Sensors ( for self monitoring )
 Effectors ( for self adjustments )
 Knowledge
 Planner ( for analyzing policies ).

Service Consolidation
 In computing, consolidation refers to when data
storage or server resources are shared among multiple
users and accessed by multiple applications.

 Consolidation aims to make more efficient use of
computer resources and prevent servers and storage
equipment from being under-utilized and taking too
much space.

 The two main types of consolidation are server
consolidation and storage consolidation.

 Server consolidation involves reducing the number
of servers and server locations within an organization.
 The intended result is more efficient use of server
resources and occupied space.
 However, this also increases the complexity of the
servers, data and applications, which may be
challenging for users.

 Server virtualization attempts to address this problem
by masking that complexity from users.

 Another option is to use blade servers, which are
actual servers in the form of modular circuit boards on
a card.
 They occupy less rack space and consume less power.

 Storage consolidation, or storage convergence, is a
method of centralizing data storage through any one
of three architectures:
 Network Attached Storage (NAS).
 Redundant Array of Independent Disks (RAID)
 Storage Area Network (SAN):

 Network Attached Storage (NAS): Dedicated storage
hard drives do not have to compete with other
computers for processing resources.

 Redundant Array of Independent Disks (RAID): Data
is located on multiple disks but appears as a single
logical drive.

 Storage Area Network (SAN): Fiber channel
technology is used to provide high throughput, data
sharing, data migration and service to many clients
(also called subscribers) over a large geographical area.
 SANS is the most sophisticated storage consolidation
method of the three.

Horizontal Scaling
 Horizontal scaling is the capability of an operation to
be scaled up to meet the demand through request and
the distribution of request across the servers as in Fig
1.3
 as demand increases the servers are scaled up.
Scalability is the capacity of a model to be enlarged to
handle the expanding volume of production in an
effective way.

 Methods of adding more resources for a particular
application falls into two broad categories, they are
Horizontal-scaling
Vertical-scaling

 This ( horizontal-scaling) is achieved by adding more
servers by using clustering and load balancing
solution.

 Numerous number of computers may be configured in
a cluster to obtain combined computing power.
Horizontal scaling is also called as scale out.

 An example of the scaling method is scaling out from
one web server system to three servers. This type of
scaling is used by the cloud computing as it is very
cheaper. In a cloud the load will be distributed across
various additional servers.

 This model created the demand in sharing data with
enormous I/O performance and particularly for the
processing of huge volume of data.

Web services
 Web services are a set of services over the web or the
technical term cloud. This a service which is “always
on”, same as in the concept of utility computing. It is a
standard way for integrating the web applications, It is
considered as the next evolution of web. Web services
converts your applications into Web application which
can be published, found and used over the internet.


 Web services communicate using open protocols and
can be used by other applications.
 These services are hardware independent, operating
system independent, programming language
independent.
 The basic platform for these services are XML and
HTTP.

 The components of web services are web service server
code, web service consumer code, SOAP, XML, WSDL
and UDDI.SOAP is a protocol for accessing web
service. It stands for Simple Object Access Protocol.
SOAP is a communication protocol for sending
messages

 XML is a markup language. It stands for eXtensible
Markup Language. The contents are encoded in xml
codes. WSDL is an XML based language. It stands
for Web Services Description Language. It is used
to describe and locate the services.

 UDDI is a directory service. It stands for Universal
Description, Discovery and Integration. UDDI is
used for storing the information about the web
services.

High Scalability Architecture
 Scalability is the ability of a model to be enlarged to
handle the growing amount of work in an effective
manner.
 The cloud resources can be rapidly scaled ( i.e.,
increased ) based on the demand which includes
storage, CPU time, memory, web service requests.
 Customers buy the services given by service provider
based on scalability, availability and performance.

 High scalability architecture is an important
parameter to select a service provider. In cloud
computing, it is important to have a High scalability
architecture that is capable of handling numerous
users on an on-demand basis.

 To achieve this, Google introduced Bigtable which has
high scalability architecture. Bigtable was developed
with very high speed, flexibility and has very high
scalability.

 A Bigtable database is petabytes(PB) in size and spans
thousands of distributed servers.
 Each bigtable is a multi dimensional table.
 The table is made up of rows and columns.
 Each cell has a time stamp. With this stamp, you can
select certain versions of a web page or delete cells that
are older than a given date and time.

 Since the tables are very large, Bigtable splits them at
row boundaries and saves them as tables.
 Each tables is about 200 MB and each server stores
about 100 tables.
 The data from a database is likely to be stored in many
different servers.

 This architecture also allows for load balancing ( i.e.,)
If one table is getting a lot of queries, it can move the
busy table to another machine that is not busy.
 Also, when a machine fills up, it can compress some
tablets freeing more drive space.

Benefits of cloud computing
 It offers various advantages in which the main
components are listed below:
 Scalability
 Simplicity
 Vendors
 Security

Scalability:
 The ability of a model to be extended to manage the
amount of work growth in an effective manner is
called scalability.
 Cloud-computing resources can be rapidly scaled
according to subscribers convenience.

 If there is a sudden necessity for more computer
resources, instead of buying new equipment we can
buy additional resources form cloud providers.
 After the endeavor is over we can stop using those
services.

Simplicity:
 In most cases cloud-computing is free to use. It is very
simple that users can easily understand which is the
biggest advantage of cloud-computing. It is possible to
get our application started instantly.

Vendors:
 The service providers are called vendors. Some of the
well known vendors
are Google, Amazon,Microsoft, IBM. These providers
offer reliable services to their customers.

Security:
 There are also some risks when using a cloud vendor.
But the reputed firms work hard to keep their
consumers data safe and secure.
 They use complex cryptographic algorithms to
authenticate users. To make it even more secure we
can encrypt our information before storing it in cloud.

Limitations in cloud
 There is not any product without a few flaws and so is
cloud computing. There are some cases where the
cloud computing may not be the best solution for
computational requirements. Such cases are called as
limitations,there are two main limitations:

 1 Sensitive Information
 2 Application development

 1 Sensitive Information :
 Storing sensitive information on the cloud is always
dangerous. Any important information about a person
or an organization is called sensitive information.

 Some of the popularly known limitations in the issue
of sensitive information are:
 Government can get the information from service
providers easily.
 In few cases, the service providers itself share our data
with marketing companies.

 The best way is to encrypt the data before storing it in
cloud or sending it to third party. Programs like PGP (
Pretty Good Privacy ) or open source True Crypt can
encrypt the file so that only the one who owns the
password can view the details stored in the uploaded
file.

 Application development :
 This is the other important limitation in cloud
computing. In some cases the applications we need
may not be available on cloud or not work as expected.

Reasons:
 And sometimes some applications will not be securely
communicated over the internet.
 In that case our data will be at risk. Thus there are only
two ways to get the desired product.

 One is, to develop their own application and other is to
approach application developer to build the desired
product for you.

Security concerns
 In cloud computing world, security is a two sided coin.
The security is very important particularly when
moving critical applications and sensitive data to
public and shared environments.

Privacy concern with a third party
 The important security concern is for privacy
considerations. That is, if third party is hosting all our
data, we do not know if it is safe or not. Everything
that is placed on cloud can be accessed by anyone.

 There are also other privacy concerns because
government can get the data that is placed on cloud
easily from organization’s servers.
 Though there are popular companies who provide
good security to keep the data safe, it can be hacked.

 The best procedure is not to perform critical tasks on a
cloud platform without extensive security.

 If it cannot be managed then it is advisable to have less
critical data on cloud.

Security level of third party
 Service providers are doing all they can to protect their
customer’s data.
 As a matter of fact, the vendors will have to make sure
that the subscriber has been fully satisfied from their
service or else the firm will not be gaining customers.

 Most of the security problems are due to loss of
control, lack of trust and multi-tenancy.

 Multi-tenancy – it is an architecture in which single
instance of a software application serves multiple
customers. Each customer is called tenant.

 These problems exist mainly in third party
management models. So there should be strong
protection measures in order to prevent the hacking of
data.

Security Benefits
 Providers do endeavor to ensure security. Cloud
provide some of the security measures ensuring the
customers data are safe:

 Centralised Data
 There are some good security traits that come
with centralizing your data, making your system more
inherently secure.

 Reduced Data Leakage:
 If the data is centralized and the various devices
used like laptop, notebook computers can access
the data, no need to backup the data.
 There is threat for theft of the handheld devices. If the
data are lost and although any security measures like
encryption is applied and it may be compromised and
the entire data may be in the hands of the thief.

 Moreover by maintaining data on the cloud,
employing strong access control, limiting the
employee downloading to only what they need to
perform a task, computing can limit the amount of
information that could be potentially be lost.

 Monitoring benefits:
 Central storage is easier to control and monitor. The
flipside is the nightmare scenario of comprehensive
data theft. If your data is maintained on a cloud, it is
easier to monitor security than have to worry about the
security of numerous servers and clients.

 The security professional figuring out smart ways to
protect and monitor access to data stored in one place
(with the benefit of situational advantage) than trying
to figure out all the places where the company data
resides.

 You can get the benefits of Thin Clients today but
Cloud Storage provides a way to centralize the data
faster and potentially cheaper. The logistical challenge
today is getting Terabytes of data to the Cloud in the
first place.

 Instant Swapover - if a server in the Cloud gets
compromised (i.e. broken into), then clone that server
at the click of a mouse and make the cloned disks
instantly available to the Cloud Forensics server.
When the swapover is performed its seamless to the
users. No need to spend time to replicate the data or
fix the breach. Abstracting the hardware allows to do it
instantly.

 Logging
 In cloud logging is improved. Logging is often an
afterthought, to solve the issues insufficient disk space
is allocated. Cloud Storage changes all this - no more
‘guessing’ how much storage you need for standard
logs.
 Contd…..

 With your logs in the Cloud you can leverage Cloud
Compute to index those logs in real-time and get the
benefit of instant search results.
 This help to Compute instances and to measure in and
scale as needed based on the logging load - meaning a
true real-time view.

 Most modern operating systems offer extended
logging in the form of a C2 audit trail. This is rarely
enabled for fear of performance degradation and log
size. Now you can ‘opt-in’ easily - if you are willing to
pay for the enhanced logging, you can do so. Granular
logging makes compliance and investigations easier.

 Secure builds
 When you developed your own network and you
have to buy third-party security software to get the
level of protection you want. With the cloud solution,
those tools can be bundled in and available to you and
you can develop your system with whatever level of
security you desire.

 Easier to test impact of security changes: this is a
big one. Spin up a copy of your production
environment, implement a security change and test
the impact at low cost, with minimal startup time.
This is a big deal and removes a major barrier to
‘doing’ security in production environments.

 Improve the state of security software
(performance)
 Drive vendors to create more efficient security
software: Billable CPU cycles get noticed. More
attention will be paid to inefficient processes; e.g.
poorly tuned security agents. Process accounting will
make a comeback as customers target ‘expensive’
processes. Security vendors that understand how to
squeeze the most performance from their software will
win.

 Security Testing
 Reduce cost of testing security: a SaaS provider only
passes on a portion of their security testing costs. It is
shared among the cloud users. The end results is that
because you are in a pool with others but you never see
the other users but you realize the lower cost for
testing.

 Even with Platform as a Service (PaaS) where your
developers get to write code, but the cloud code –
scanning tools check for security weakness.

Government Policies:
 To implement the policy, Government of India has
made an initial step “GI Cloud” which has been coined
as ‘Meghraj’. The focus of this initiative is to accelerate
delivery of e-services in the country while optimizing
the expenditure of the Government.

 Meity( Ministry of Electronics and Information
Technology) has empanelled the following companies
for providing cloud computing services to government
departments :
 Microsoft Corp.,
 Hewlett Packard,
 IBM India ,
 Tata Communications,
 Bharat Sanchar Nigam Ltd (BSNL),
 Net Magic IT Services,
 Sify Technologies and
 CtrlS Data Centers.

cloud computing basics

More Related Content

What's hot (20)

Similar to cloud computing basics (20)

Recently uploaded (20)

cloud computing basics