CLOUD COMPUTING AND GRID COMPUTI

							   INTRODUCTION
   GRID COMPUTING
   CLOUD COMPUTING
   COMPARISION BETWEEN GRID AND CLOUD
    COMPUTING
   CONCLUSION
 VISION OF CLOUD COMPUTING AND GRID
 COMPUTING IS SAME

1)   Reduce the cost of computing
2)   Increase reliability
3)   Increase flexibility
The term Grid computing was coined to
describe technologies that would allow
consumers to obtain computing power on
demand

Grid computing is the combination of
computer resources from multiple
administrative domains for a common goal

Grid computing is the back bone of the cloud
computing.
   Cloud computing is internet based
    computing, whereby shared resources,
    software and information are provided to
    computers and other devices on-demand,
    like a public utility.
   Cloud computing is different grid computing
    in several ways
      1)It is massively scalable.
      2)The services can be dynamically
         configured.
    The vision for Clouds and Grids are similar ,
    technologies used may differ and the two
    communities are struggling with many of the
    same issues. They are

 Business Model
 Architecture
 Resource Management
 Programming Model
 Application Model
 Security Model
In Grid computing business model for s/w has been a
 one-time payment for unlimited use.

The business model for grids is project oriented.

TeraGrid operates in this fashion.

In Cloud based business model , the customer will pay
the provider on computation basis. Example Amazon
uses Compute cloud EC2 and Data Cloud S3. Former is
charged based on per instance hour and later by per
GB-Month of storage used.
Grids started of in mid-90’s to address large
scale computational problems. Major
motivation was these high performance
computing resources were expensive in order
to over come this grid computing is
introduced.
  Grid protocol architecture mainly consists of
  five layers
 Fabric layer: Consists of raw hardware level
    resources.
 Connectivity Layer: Defines core
  communication and authentication protocols
  for easy and secure network transactions.
 Resource layer: Defines protocols for the
  publication .GRAM(Grid Resource Access and
  Management) protocol is used in this layer.
   Collective layer: This layer captures
    interactions across collections of resources
    Directory services such as MDS( monitoring
    and Discovery Services) allows for the
    monitoring.
   Application Layer: Comprises whatever user
    applications built on top of above protocols.
Clouds are developed to address internet-
scale computing problems .Clouds are
referred to as a large pool of computing and
storage resources , which are accessed by
standard protocols.
   The architecture Fabric layer deals with raw
    hardware like computing resources
   Unified Resource layer contains resources
    that they can be exposed to upper layer and
    end users
   Platform layer adds on collection of
    specialized tools , middle ware .
   Application layer contains the applications
    that would run in the clouds
    Resource management covers several topics
    like:
•   Compute Model
•   Data Model
•   Virtualization
•   Monitoring
•   provenance
Compute Model: Most Grids use a batch-
scheduled compute model, in which local
resource manages the compute resources of
grid site. The jobs requested by users are
queued according to availability of resources.

Cloud computing model will likely look very
different ,with resources in the cloud being
shared
Data Model: Data grids are specially designed to
tackle data in intensive applications in grid
environments.
Concept of virtual data plays a vital role in grid
computing.
Location transparency is another area where data
can requested without regard to track data
location.
Representation transparency where data can be
consumed or produced no matter of actual
physical data
Data Model: many people think that future
internet computing will be towards Cloud
Computing , Where the internet computing
will be centralized around Data, Clouding
Computing as well as Client Computing.

If this come true then the critical role of
cloud computing goes without any use and
users might not share all the information due
to some security issues
 Virtualization: Due to virtualization clouds are able to run millions
  of users applications.
  virtualization provides necessary abstraction such that the
  underlying fabric can be unified as pool of resources
 virtualization enables each application to be encapsulated
 Virtualization allows quick recovery from unplanned outages , as
  virtual environments can be backed up .
 By virtualization common resources can be cached and reused.
 Virtualization will configure resource requirements for various
  applications.

 The Grids do not relay on virtualization as much as clouds
  do, but that is due to policy and having each individual
 organization maintain full control of resources.
   Monitoring: Due to concept of virtualization
    monitoring the resources is a challenge. In
    cloud different level of services are offered to
    different users especially through ( SASS and
    PASS ) level as the resources are opaque to
    the users.

     Girds in general have different trust model in
    which users access resources via there
    identity .So monitoring is quite easy in grid
    computing
   Provenance: Provenance refers to derivation
    history of data.
    In Grids provenance management has been in
    general built into workflow system from early
    pioneers like CHIMERA

    But Provenance is still as unexplored area in
    cloud computing in which there is a need to
    deal with even more challenging issues such
    as tracking data production across different
    service providers
Grids use the traditional parallel and distributed
Environments. MPI(Message passing Interface)
MapReduce are the two most commonly used
parallel Programming's in GRIDS

Java Script ,PHP, Python are commonly used
programming languages in Clouds because
they take outputs from one application and
transform it into another. Google App engine
 uses modified python runtime and python
 scripting language
Grids generally support many different kinds
of application. Applications such as HPC(
executing tightly coupled parallel jobs),
HTC(executing loosely coupled natured jobs).

Cloud computing could not support HPC(
High Performance Computing) but can
support HTC(High Throughput Computing)
Security model of clouds seems to be relatively
simpler and less secure since the cloud
infrastructure depends on Web forms which is
less secure and new users could use clouds
relatively easily and almost instantly.

Grids are stricter about its security where
information about the new accounts requires
person to person rather than verification from
sponsoring person. It is time consuming.
We can conclude that clouds and grids share
a lot commonality in the vision , architecture
and technology , but they also differ in
various aspects such as security ,
programming model, compute model,
business model, data model , applications
and abstractions
We have seen the different challenges and
opportunities in both the fields
References:

•   Ian Foster, Ioan Raicu “ University of Chicago”
•   Ian Foster “ Department of Computer science
    University of Chicago”
•   Ian Foster “ Math & computer science,
    University of Chicago”
•   Yong Zhao “ Microsoft corporation ,
    Redmond”
THANK YOU
?