“Grid” computing is to reach a common goal from
multiple location by using a collection of computer resource is called the Grid
Computing, the computers on the network can work on a task together, thus
functioning as a supercomputer. Grid computing is the combination of computer
resources from multiple administrative domains applied to a common task,
usually to a scientific, technical or business problem that requires a great
number of computer processing cycles or the need to process large amounts of
data. It is a type of parallel and distributed system that enables the sharing,
selection, and aggregation of geographically distributed autonomousresources
dynamically at runtime depending on theiravailability, capability, performance,
cost and usersquality-of-service requirements 13.
It is a shared collection of reliable &
unreliable resources andinteractively communicating researchers of
differentvirtual organizations (doctors, biologists). This systemcontrols and
coordinates the integrity of the Grid bybalancing the usage of reliable and
unreliable resourcesamong its participants providing better quality ofservice.
It can also be denoted by large-scale clustercomputing, as well as a form of
network-distributedparallel processing. It is a collection of servers that
areclustered together to attack a single problem 73.
1.2 Reasons for Grid Computing
A well designed grid can improve performance if
some applications running on the grid are idle when resources are needed for
other applications. Computational approaches to problem solving have proven
their worth in almost every field of human endeavour. But, there are some
challenging problems that increase computer’s ability to solve them, therefore,
one important factor is that the much of the computing environment isincomplete
for such computationally sophisticated purposes. While today’s PC is faster
than the supercomputer of 10 years ago, it is still far from adequate for
predicting the outcome of complex actions or selecting from among many choices.
That, after all, is why super computing environments have continued to evolve.
1.3 History of Grid
In the early 1970’s when computers were
first linked by networks, the idea of harnessing unused CPU cycles was born. A
few early experiments with distributed computing including a pair of programs
called Creeper and Reaper ran on the Internet’s predecessor, the ARPAnet.
In 1973, the Xerox Palo Alto Research
Center (PARC) installed the first Ethernet network and the first full-fledged
distributed computing effort was underway. Scientists John F. Shoch and Jon A.
Hupp created a worm, as they called it, and envisioned it moving from machine
to machine using idle resources for beneficial purposes.
In another effort, Richard Crandall started
putting idle, networked NeXT™ computers to work. Crandall installed software
that allowed the machines, when not in use, to perform computations and to
combine efforts with other machines on the network. In 1991, Zilla won the
ComputerWorld Smithsonian Award for Science.
In another project started in 1995, I-WAY
(Information Wide Area Year) was an experimental high-performance network
linking many high-performance computers and advanced visualization
environments. The idea was not to create a new network, but to integrate
existing high bandwidth networks. So I-WAY combine resources at multiple
supercomputing centers. It was designed to provide an infrastructure to a range
of high performance applications. The reason why I-WAY is so important is that
it is the startup of Globus, which is the de facto standard for developing grid
1.3.1 The Role of Internet in
Distributed computing jump
to a global level with the maturation of the Internet in the 1990’s. Two
projects in particular have proven that the concept works extremely well even
better than many experts had introduce.
First of these
revolutionary projects used thousands of independently owned computers across
the Internet to crack encryption codes. This project was
called distributed.net affectionately known as “dnet”.
The second one is the
“[email protected]” project which was the most successful and popular of distributed
computing projects in history. Over three million peoplehave installed the
[email protected] software agent since the project started in May 1999. This project completely
proved that distributed computing could sharpen computing project results while
managing project costs.
Since 1999 grid
computing has further developed to encompass more of IT than just computers and
data. Grid enables a loosely-coupled, service-based IT environment. It is the
broad spectrum of IT “resources” that can be a “service”
that elevates Grid beyond just scientific computing. 9
is the collection of resources provided by multiple systems in the network thus
grid provide protocols and services at five different layers as identified in
the Grid protocol architecture. Following are five layers in grid architecture.
Grids provide access to different
resource types such as compute, storage and network resource, code repository,
etc. Grids usually rely on existing fabric components, for instance, local
2.2 Connectivity Protocols
It defines core communication and
authentication protocols for easy and secure network transactions. The GSI
(Grid Security Infrastructure) protocol underlies every Grid transaction.4
It defines protocols for the publication, discovery,
negotiation, monitoring, accounting and payment of sharing operations on
individual resources. The GRAM (Grid Resource Access and Management) protocol
is used for allocation of computational resources and for monitoring and
control of computation
On those resources, and Grid FTP for data access and
high-speed data transfer.4
2.4 Collective Services
The Collective layer captures interactions across
collections of resources, directory services such as MDS (Monitoring and
Discovery Service) allows for the monitoring and discovery of VO resources.4
2.5 User Application
It comprises whatever user applications built on top
of the above protocols and APIs and operate in VO environments. Two examples
are Grid workflow systems, and Grid portals.4
3 Grid Middleware
Grid Middleware are software stacks designed to
present disparate compute and data resources in a uniform manner, such that
these resources can be accessed remotely by client software without needing to
know a priori the systems’ configurations. Currently The CSAR machines support
basic Globus and UNICORE access.
Globus is a middleware constructed from a
number of components which make up a toolkit. This toolkit provides client,
server and development components for the three Globus “pillars” of
Grid Computing: Resource management, Information Management and Data
UNICORE (UNiform Interface to COmpute
REsources) provides a science and engineering grid combining resources of
supercomputer centres and making them available through the Internet. Strong
authentication is performed in a consistent and transparent manner, and the differences
between platforms are hidden from the user thus creating a seamless HPC portal
for accessing supercomputers, compiling and running applications, and
transferring input/output data.
Access to both flavours of middleware is based
upon user owned X509 certificates (such as those issued by the UK eScience
Certificate Authority). UNICORE makes use of these certificates directly
whereas Globus allows the use of GSI (Grid Security Infrastructure) proxy
impersonation certificates. Both methods allow for single sign-on to varying
3.2 Middleware Applications
The grid middleware allows the users to submit their
program requests to execute jobs, or computations, to their grid. This is simple,
because it allows the jobs to be executed anywhere on computer network. The
software components, like resource brokers, pass the job to the node, or
network point, that is the most appropriate. Middleware is also useful for the
security functions, since it allows the users to give themselves authorization
through standard certificate infrastructure.
In the business world, time is money and grid
middleware can save companies precious time. It offers businesses
interoperability and seamlessly sends the jobs to the appropriate network
point. Grid middleware is what makes everything on a computer network to run
well and to ensure that a company is working to its fullest capacity.10
When searching for a supplier for grid middleware,
it’s important to search for a company that has a long history of satisfied
customers. Apprenda is a company that knows the importance of smooth operations
and has created many applications suited to the needs of today’s companies,
including grid middleware.
4 Grid security
In grid the security is defined in the
resource layer of grid architecture. The resource which is use in grid and the task
which is to be solved are both important and valuable. The security problems in
grid environment are complex because resources are located in different
administrative domains with each resource potential having its own policies and
procedures. The security service is a processing or communication service provided
by a system to give a specific kind of protection to system resources. Security
services implement security plans and are implemented by security mechanism.
Security concerns are difficult by the fact that there are different
requirements by users, resource owners, designers who are creating or adapting
their current products and tools to take pro of the grid technology. Following
are the main Issues in grid security.
These issues address the pointsrelated to the architecture of the
grid. The users of the grid are concerned about the data powdered by the grid
and hence there is a need to protect the data confidentiality and integrity as
well as the user validation. Architecture level issues may include issues like
information security, authorization and service level security which
destabilize the whole system and hence an architectural level solution is
needed to prevent those.
4.2 Infrastructure Related Issues
These issues are related to the network and host which are found in
the grid infrastructure. Host level security issues are individual’s issues
that make a host apprehensive about affiliating itself to the grid system.
The issues that are related to the infrastructure may include data
protection, job starvation, and host accessibility. The infrastructure related
issues are of two types: host security issues and network security issues. The
host level security issues are introduce above. The network security issues
arise mainly due to the heterogeneity and high speed requirements of many grid
applications. Many of the grid network issues are active areas of research and
are most developed in labs and not yet commercialized.
4.3 Management Related Issues
The third set of issues relate to the management of the grid. Managing
pass is absolutely important in grid systems because nature of the grid frame
and applications is mixed. Like any distributed system, managing belief is also
serious and falls below the purview of management related issues. The different
management issues are credential management, trust management and monitoring
related issues. Management of credentials is very important in grid context as
there are multiple different systems which varied credentials to access them.
Management of trust is very difficult in a dynamic grid scenario where grid
nodes and users join and leave the system. Monitoring of resources consists of
different stages such as collection, processing, transmission, storage and
presentation of the data.
5 Challenges of grid computing
A lot of heterogeneous hardware is used in order to
create the Grid and, in addition, these devicesare not managed by only one
person but by different system administrators in each of the companies. Grid
follows the challenges that need to be resolved to harness the full power of
No clear standard
Grid computing uses various standards, but all grids
are not use same standards. Example all grid operating system such as Linux,
Apache and My SQL are using WSRF, UDDI,WWW, SOAP and XML standards. Oracle 10g
enterprise implement without WSRF. IBM develops the Grid middleware based on
J2EE. We cannot use different OS at the same machine in the same time in grid
5.2 Distributed computing Vs Grid computing
Grid computing involves dynamic virtual organization,
resource sharing and peer to peer computing. The
Grid intends to make access to computing power,
scientific data repositories and experimental facilities
As easy as the Web makes access to information. Same
all facilities provide the grid computing. So it is a challenge for grid
5.3 Lack of grid enabled software
The software, which are enabled the grid computing
are less, It has limited software on Grid. Much software has not copyright
issues and source code of license. It is need for more company developing
grid-enabled version, need more developers on grid development and need to
develop open source software.
5.4 Sharing Resources between Various Types of
Grid used for sharing resource from various sites
and grid hosts. It handles a massive amount of data as a grid platform. A lot
of sites and multiple servers gathered there it is so complex infrastructure.
It provides difficulty for hardware resource sharing within virtual organization.
Many institutes and organizations used grid
computing. It distributes the resources on largegeographicallydistributed
environments and accessesthe heterogeneous devices andmachines. So it is a
major challenge tomanage the administration of the grid computing.
6 Current Grid Projects
Grid Computing have a lots of active projects some of the projects
6.1 Drug Discovery @Home (BOINC)
Help Drug Discovery @Home “model the behavior of leading
compounds that could be developed into new medicines.” The project
“is in an early alpha phase and does not have a formal relationship with
academia or the pharmaceutical industry.”
The project uses a BOINC-based client. See the BOINC platform
information for the latest version of the BOINC client. See the project’s
applications page for a list of its clients, their latest versions, and the
platforms they are available for.
6.2 RNA World (BOINC)
The project’s first goal is to systematically identify all known
RNA family members in all organisms currently known and to make the information
available to the public. The project will also run sub-projects submitted by
The project uses a BOINC-based client. See the BOINC platform
information for the latest version of the BOINC client.
6.3 Spinhenge @home (BOINC)
“Berkeley Open Infrastructure for Network Computing (BOINC)
is a platform for projects, like distributed.net and SETI @home that use
millions of volunteer computers as a parallel supercomputer.” It provides
tools create and manage distributed computing projects, to create project
websites, to automate the translation of project websites, and tools for
BOINC-based project participants to easily manage their multiple accounts.
Source code is available for the platform, and interested C++ developers are
encouraged to help develop the platform code
6.4 Distributed.net RC5
Distributed. Net’s RC5 project crack data encryption schemes
(legally). distributed.net, which began in 1997, was the first well-known
public distributed computing project and the RC5 project (specifically the
RC5-32/12/7 (56-bit) project) was its first project. The RC5 project uses a
brute-force search technique to find an encryption key, testing each possible
key until it finds one which successfully decrypts a test message.
6.5 Enigma @Home (BOINC)
The project uses a BOINC-based client. See the BOINC platform
information for the latest version of the BOINC client. Version 5.17 of the project’s
Enigma 0.76 application is available for Windows as of September 11, 2007.
Version 5.20 of the application is available for Linux as of September 25,
2007. Version 5.22 of the project’s Enigma 0.76b application is available for
Windows and Linux as of August 24, 2008