[attachment=10507]
INTRODUCTION
1.1 Overview of the project
Grid Computing, in general terms, means applying the resources of many computers in a network to execute a process that requires a great number of computer processing cycles or access to large amounts of data. Ideal Grid is still a virtual concept when it goes well beyond simple communication between computers, and aims ultimately to turn the global network of computers into one vast computational resource. An Ideal Grid 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 intensive problems.
Infinity, Grid Computational Framework, limits the design to specific domain of network and types of resources to be used for sharing and selecting, still maintaining the integrity of an abstract Grid. Framework is based on Middleware Architecture which provides most of the major functionality provided by the software. Middleware runs in all the system that are a part of the grid and this component is the one with which the grid directly interacts. The architecture is completely peer-to-peer with each system in the grid capable of submitting the job to the Middleware independently. Dot net is used for implementation of the framework designed due to its effective server application and better network application.
Finally the implementation language due to large set of functionalities it provides to make the work more sophisticated and effective in nature. Also, platform independent, providing option opened for extending the software to work independent of the platform.
1.2 Need and Scope of this project
The purpose of the project, Infinity, is to design and develop a Grid Communicator Framework, exploiting under-utilized computing resources over a network. Grid will provide remote execution of jobs that require very high computational power in terms of CPU and memory space .Software will search for the best available system in the grid by collecting the system status at that instance of time and determine the most feasible system that can used to execute the job .The details of the design and implementation of the framework is described later in the report. The basic features employed in the Grid Computational Framework are as follows:-
• Remote execution of the jobs requiring high CPU and memory capacity in a connected network of LAN.
• Harnessing massive power of unutilized resources scattered all over the network (LAN).
• Registering of the services so that once a job has been submitted and executed remotely on the grid ,user would have to just pass the argument when next time he wishes to execute the same program.
• Provide a graphical interface for running the job in the grid.
1.3 Advantages of Proposed Technology
• Remote execution of the jobs requiring high CPU and memory capacity in a connected network of LAN.
• Harnessing massive power of unutilized resources scattered all over the network .
• Grid environments are much more modular and don't have single points of failure. If one of the servers/desktops within the grid fails there are plenty of other resources able to pick the load. Jobs can automatically restart if a failure occurs.
• Jobs can be executed in parallel speeding performance. Grid environments are extremely well suited to run jobs that can be split into smaller chunks and run concurrently on many nodes. Using things like MPI will allow message passing to occur among computer resources.
• Registering of the services so that once a job has been submitted and executed remotely on the grid, user would have to just pass the argument when next time he wishes to execute the same program.
• Provide a graphical interface for running the job in the grid.
1.4 Drawback of Existing System
• Need to buy large six figure SMP servers for applications that can be split up and farmed out to smaller commodity type servers.
• Much more efficient use of idle resources. Jobs can be farmed out to idle servers or even idle desktops. Many of these resources sit idle especially during off business hours.
1.5 Software Requirements
• Operating System: Windows XP
• Presentation layer: Win app (.net)
• Language: C#
1.6 Hardware Requirements
• Processor: P4 and Above
• HDD: 5 GB
• RAM: 256Mb and Above
1.7 Motivation
This section briefly presents the motivational factors behind deciding upon Infinity as the to-be-pursued project. The principal motivational factors that inspired us were:
• Grid computing is an emerging technology that enables large-scale resource sharing and execution of jobs remotely. The future prospect in present area is very high and a knowledge about grid computing at this juncture of time will enable us to explore more of this technology in subsequent areas.
• Much of the computational power in our LAN is being wasted as most of time only title amount of CPU Cycles and memory space are utilized. CPU and memory remains free for a major portion of time leaving out some of rare occasion. These resources can be harnessed and provide it to any system that requires it more and asks for it.
• Even though many of the developers have designed the framework for Grid Computing, lot of areas are yet to be explored. The dream of having a virtual Grid of computing devices that are not only connected to each other, rather creating a environment where users can execute any job relieved from the burden of worrying about CPU cycles and memory, is still unfulfilled. The project is an initiative to reach the dream in the constrained domain of LAN and first step towards many more projects to come in this field.
The above mentioned points promoted to start from the scratch and develop a Grid Computing Framework of our own.