GRID COMPUTING

Jenifer.s1, Mallika.P2
Department Of Informatiom Technology

Park College of Engineering & Technology

[email protected]

[email protected]

This paper aims to present the state-of-the-art concepts of Grid computing.

A set of general principles, services and design criteria that can be
ABSTRACT--- Mankind is right in the middle of another evolutionary followed in the Grid construction are given. One of the Grid application
technological transition which once more will change the way we do project Legion is taken up. We conclude with future trends in this yet to be
things. And, you guessed right, it has to do with the Internet. It’s conquered technology.
called "The Grid", which means the infrastructure for the Advanced I .INTRODUCTION
Web, for computing, collaboration and communication. The popularity of the Internet as well as the availability of powerful
Grid computing, most simply stated, is distributed computing taken to computers and high-speed network technologies as low-cost commodity
the next evolutionary level. The goal is to create the illusion of a simple components is changing the way we use computers today. These
yet large and powerful self managing virtual computer out of a large technology opportunities have led to the possibility of using distributed
collection of connected heterogeneous systems sharing various computers as a single, unified computing resource, leading to what is
combinations of resources. popularly known as Grid computing.
 The term Grid is chosen as an analogy to a power Grid that collaborative engineering, data exploration, high-throughput computing,
provides consistent, pervasive, dependable, transparent access to electricity and distributed supercomputing.
irrespective of its source. This new approach to network computing is
known by several names, such as metacomputing, scalable computing,
global computing, Internet computing, and more recently peer to peer
(P2P) computing.
Grids enable the sharing, selection, and aggregation of a wide variety of
resources including supercomputers, storage systems, data sources, and
specialized devices that are geographically distributed and owned by
different organizations for solving large-scale computational and data Grid Computing: A Conceptual View
intensive problems in science, engineering, and commerce. Thus creating
virtual organizations and enterprises as envisioned in as a temporary
II. SERVICES OFFERED BY GRID
alliance of enterprises or organizations that come together to share
resources and skills, core competencies, or resources in order to better A Grid can be viewed as a seamless, integrated computational and
respond to business opportunities or large-scale application processing collaborative environment and a high-level view of activities within the
requirements, and whose cooperation is supported by computer networks. Grid as shown in Figure. The users interact with the Grid resource broker
The concept of Grid computing started as a project to link geographically to solve problems, which in turn performs resource discovery, scheduling,
dispersed supercomputers, but now it has grown far beyond its original and the processing of application jobs on the distributed Grid resources.
intent. The Grid infrastructure can benefit many applications, including From the end-user point of view, Grids can be used to provide the
following types of services.
  Application services: These are concerned with application
 Computational services: These are concerned with providing secure management and providing access to remote software and libraries
services for executing application jobs on distributed computational transparently. The emerging technologies such as Web services are
resources individually or collectively. Resources brokers provide expected to play a leading role in defining application services.
the services for collective use of distributed resources. A Grid They build on computational and data services provided by the
providing computational services is often called a computational Grid. An example system that can be used to develop such services
Grid. Some examples of computational Grids are: NASA IPG, the is NetSolve.
World Wide Grid, and the NSF TeraGrid.  Information services: These are concerned with the extraction and
presentation of data with meaning by using the services of
 Data services: These are concerned with proving secure access to computational, data, and/or application services. The low-level
distributed datasets and their management. To provide a scalable details handled by this are the way that information is represented,
storage and access to the data sets, they may be replicated, stored, accessed, shared, and maintained. Given its key role in
catalogued, and even different datasets stored in different locations many scientific endeavors, the Web is the obvious point of
to create an illusion of mass storage. The processing of datasets is departure for this level.
carried out using computational Grid services and such a  Knowledge services: These are concerned with the way that
combination is commonly called data Grids. Sample applications knowledge is acquired, used, retrieved, published, and maintained
that need such services for management, sharing, and processing of to assist users in achieving their particular goals and objectives.
large datasets are high-energy physics and accessing distributed Knowledge is understood as information applied to achieve a goal,
chemical databases for drug design. solve a problem, or execute a decision. An example of this is data
mining for automatically building a new knowledge.
 applications that require a large number of geographically located
resources must be designed to be latency and bandwidth tolerant.
IV. GRID CONSTRUCTION: GENERAL PRINCIPLES
 Dynamicity or adaptability: In a Grid, resource failure is the rule
This section briefly highlights some of the general principles that underlie
rather than the exception. In fact, with so many resources in a Grid,
the construction of the Grid. In particular, the idealized design features that
the probability of some resource failing is high. Resource managers
are required by a Grid to provide users with a seamless computing
or applications must tailor their behavior dynamically and use the
environment are discussed. Four main aspects characterize a Grid.
available resources and services efficiently and effectively.
 Multiple administrative domains and autonomy: Grid resources are
geographically distributed across multiple administrative domains
The steps necessary to realize a Grid include:
and owned by different organizations. The autonomy of resource
 The integration of individual software and hardware components
owners needs to be honored along with their local resource
into a combined networked resource (e.g. a single system image
management and usage policies.
cluster);
 Heterogeneity: A Grid involves a multiplicity of resources that are
 The deployment of:
heterogeneous in nature and will encompass a vast range of
– low-level middleware to provide a secure and transparent
technologies.
access
to resources;
 Scalability: A Grid might grow from a few integrated resources to
– user-level middleware and tools for application
millions. This raises the problem of potential performance
development and the
degradation as the size of Grids increases. Consequently,
aggregation of distributed resources;
 The development and optimization of distributed applications to The components that are necessary to form a Grid (shown in Figure) are as
follows.
take advantage of the available resources and infrastructure.
 Grid fabric: This consists of all the globally distributed resources
that are accessible from anywhere on the Internet. These resources
could be computers (such as PCs or Symmetric Multi-Processors)
running a variety of operating systems (such as UNIX or
Windows), storage devices, databases, and special scientific
instruments such as a radio telescope or particular heat sensor.
 Core Grid middleware: This offers core services such as remote
process management, co-allocation of resources, storage access,
information registration and discovery, security, and aspects of
Quality of Service (QoS) such as resource reservation and trading.
 User-level Grid middleware: This includes application
development environments, programming tools and resource
brokers for managing resources and scheduling application tasks
for execution on global resources.
 Grid applications and portals: Grid applications are typically
developed using Grid-enabled languages and utilities such as HPC+
Grid Architecture and Components + or MPI. An example application, such as parameter simulation or
a grand-challenge problem, would require computational power,
 access to remote data sets, and may need to interact with scientific system. Legion defines an API for object interaction, but not the
instruments. Grid portals offer Web-enabled application services, programming language or communication protocol.
where users can submit and collect results for their jobs on remote  Classes manage their instances: Every Legion object is defined and
resources through the Web. managed by its own active class object. Class objects are given
system-level capabilities; they can create new instances, schedule
Legion An Example Grid Project: them for execution, activate or deactivate an object, as well as
There are many international Grid projects worldwide, which are provide state information to client objects.
hierarchically categorized as integrated Grid systems, core middleware,  Users can define their own classes: As in other object-oriented
user-level middleware, and applications/application driven efforts. Now we systems users can override or redefine the functionality of a class.
look briefly about a project which is an object-based metasystem This feature allows functionality to be added or removed to meet a
developed at the University of Virginia. user’s needs.

Legion provides the software infrastructure so that a system of Legion core objects support the basic services needed by the metasystem.
heterogeneous, geographically distributed, high-performance machines can The Legion system supports the following set of core object types.
interact seamlessly. Legion attempts to provide users, at their workstations,  Classes and metaclasses: Classes can be considered managers and
with a single, coherent, virtual machine. policy makers. Metaclasses are classes of classes.
In the Legion system the following apply.  Host objects: Host objects are abstractions of processing resources;
 Everything is an object: Objects represent all hardware and they may represent a single processor or multiple hosts and
software components. Each object is an active process that processors.
responds to method invocations from other objects within the
  Vault objects: Vault objects represent persistent storage, but only
for the purpose of maintaining the state of Object Persistent There are currently a large number of projects and a diverse range of new
Representation (OPR). and emerging Grid developmental approaches being pursued. These
 Implementation objects and caches: Implementation objects hide systems range from Grid frameworks to application testbeds, and from
the storage details of object implementations and can be thought of collaborative environments to batch submission mechanisms. It is difficult
as equivalent to executable files in UNIX. Implementation cache to predict the future in a field such as information technology where the
objects provide objects with a cache of frequently used data. technological advances are moving very rapidly. Hence, it is not an easy

 Binding agents: A binding agent maps object IDs to physical task to forecast what will become the ‘dominant’ Grid approach. Windows

addresses. Binding agents can cache bindings and organize of opportunity for ideas and products seem to open and close in

themselves into hierarchies and software combining trees. the ‘blink of an eye’. However, some trends are evident. One of those is

 Context objects and context spaces: Context objects map context growing interest in the use of Java and Web services for network

names to Legion object IDs, allowing users to name objects with computing.

arbitrary-length string names. Context spaces consist of directed

graphs of context objects that name and organize information.
VI .BIBILIOGRAPHY
V. CONCLUSION:
The Grid is analogous to the electricity (power) Grid and the vision are to 1. Foster I, Kesselman C The Grid: Blueprint for a Future Computing

offer (almost) dependable, consistent, pervasive, and inexpensive access to Infrastructure. Morgan Kaufmann: San Francisco, CA, 1999.

resources irrespective of their location for physical existence and their 2. www.cs.mu.oz.au

location for access. 3. www.gridbus.org