AddThis

Bookmark and Share

Tuesday 10 August 2010

GRID COMPUTING

SEMINAR ON GRID COMPUTING
Grid Computing is a technique in which the idle systems in the Network and their “wasted” CPU cycles can be efficiently used by uniting pools of servers, storage systems and networks into a single large virtual system for resource sharing dynamically at runtime.
- High performance computer clusters.
-share application, data and computing resources.


IMPORTANCE OF GRID COMPUTING
Flexible, Secure, Coordinated resource sharing.
Virtualization of distributed computing resources.
Give worldwide access to a network of distributed resources.


GRID REQUIREMENTS
Security
Resource Management
Data Management
Information Services
Fault Detection
Portability


TYPES OF GRID
Computational Grid
-computing power
Scavenging Grid
-desktop machines
Data Grid
-data access across multiple organizations


ARCHITECTURAL OVERVIEW
- Grid’s computer can be thousands of miles apart and connected with internet networking technologies.
- Grids can share processors and drive space.


Fabric : Provides resources to which shared access is mediated by grid protocols.
Connectivity : Provides authentication solutions.
Resources : Connectivity layer, communication and authentication protocols.
Collective : Coordinates multiple resources.
Application : Constructed by calling upon services defined at any layer.

GRID COMPONENTS
In a world-wide Grid environment, capabilities that the infrastructure needs to support include:
Remote storage
Publication of datasets
Security
Uniform access to remote resources
Publication of services and access cost
Composition of distributed applications
Discovery of suitable datasets
Discovery of suitable computational resources
Mapping and Scheduling of jobs
Submission, monitoring, steering of jobs execution
Movement of code
Enforcement of quality
Metering and accounting

GRID LAYERS
Grid Fabric layer
Core Grid middleware
User-level Grid middleware
Grid application and protocols
OPERATIONAL FLOW FROM USER’S PERSPECTIVE
- Installing Core Gridmiddleware
- Resource brokering and application deployment services

COMPONENT INTERACTION
- Distributed application
- Grid resource broker
- Grid information service
- Grid market directory
- Broker identifies the list of computational resources
- Executes the job and returns results
- Metering system passes the resource information to the accounting system
- Accounting system reports resource share allocation to the user

PROBLEM AND PROMISES
PROBLEMS
- Coordinated resource sharing and problem solving in dynamic, institutional organizations
- Improving distributed management
- Improving the availability of data
- Providing researchers with a uniform user friendly environment
PROMISES
- Grid utilizes the idle time
- Its ability to make more cost-effective use of resources
- To solve problems that can’t be approached without any enormous amount of computing power.



CONCLUSION
- Grid Computing is becoming the platform for next generation escience experiments
- By Intranet Grid it is very easy to download multiple files

0 comments:

Post a Comment