WELCOMEWELCOME

GRID GRID COMPUTINGCOMPUTING

NAME:Navpreet Kaur

BRANCH: CSE(Evening Shift)

ROLL NO.: 115361

What is Grid Computing?What is Grid Computing? Cousins of grid computing.Cousins of grid computing. Methods of grid computing.Methods of grid computing. Who Needs It?Who Needs It? Grid UsersGrid Users Some highly visible grids.Some highly visible grids. Using the grid.Using the grid.

What is grid computing?What is grid computing?

Grid computing involves connecting Grid computing involves connecting geographically remote computers into a single geographically remote computers into a single network to create a virtual supercomputer by network to create a virtual supercomputer by combining the computational power of all combining the computational power of all computers on grid.computers on grid.

COMPUTATIONAL GRIDS COMPUTATIONAL GRIDS

HomogeneousHomogeneous HeterogeneousHeterogeneous

A network of geographically distributed A network of geographically distributed resources including computers, peripherals, resources including computers, peripherals, switches, instruments, and data.switches, instruments, and data.

Each user should have a single login account Each user should have a single login account to access all resources.to access all resources.

Resources may be owned by diverse Resources may be owned by diverse organizationsorganizations

Cousins of Grid ComputingCousins of Grid Computing

Distributed ComputingDistributed Computing Peer-to-Peer Computing etc.Peer-to-Peer Computing etc. Distributed ComputingDistributed Computing Distributed computing is most often Distributed computing is most often

concerned with distributing the load of a concerned with distributing the load of a program across two or more processesprogram across two or more processes

PEER2PEER ComputingPEER2PEER Computing

Sharing of computer resources and services by Sharing of computer resources and services by direct exchange between systems.direct exchange between systems.

Computers can act as clients or servers Computers can act as clients or servers depending on what role is most efficient for depending on what role is most efficient for the network.the network.

Methods of Grid ComputingMethods of Grid Computing

Distributed SupercomputingDistributed Supercomputing High-Throughput ComputingHigh-Throughput Computing On-Demand ComputingOn-Demand Computing Data-Intensive ComputingData-Intensive Computing Collaborative ComputingCollaborative Computing Logistical NetworkingLogistical Networking

Distributed SupercomputingDistributed Supercomputing

Combining multiple high-capacity resources Combining multiple high-capacity resources on a computational grid into a single, virtual on a computational grid into a single, virtual distributed supercomputer.distributed supercomputer.

Tackle problems that cannot be solved on a Tackle problems that cannot be solved on a single system.single system.

High-Throughput ComputingHigh-Throughput Computing

Uses the grid to schedule large numbers of Uses the grid to schedule large numbers of loosely coupled or independent tasks, with the loosely coupled or independent tasks, with the goal of putting unused processor cycles to goal of putting unused processor cycles to work.work.

On-Demand ComputingOn-Demand Computing

Uses grid capabilities to meet short-term Uses grid capabilities to meet short-term requirements for resources that are not locally requirements for resources that are not locally accessible.accessible.

Data-Intensive ComputingData-Intensive Computing

The focus is on synthesizing new information The focus is on synthesizing new information from data that is maintained in geographically from data that is maintained in geographically distributed repositories, digital libraries, and distributed repositories, digital libraries, and databases.databases.

Particularly useful for distributed data mining.Particularly useful for distributed data mining.

Collaborative ComputingCollaborative Computing

Concerned primarily with enabling and Concerned primarily with enabling and enhancing human-to-human interactions. enhancing human-to-human interactions.

Applications are often structured in terms of a Applications are often structured in terms of a virtual shared space.virtual shared space.

Logistical NetworkingLogistical Networking

Global scheduling and optimization of data Global scheduling and optimization of data movement.movement.

Contrasts with traditional networking, which does not Contrasts with traditional networking, which does not explicitly model storage resources in the network. explicitly model storage resources in the network.

Called "logistical" because of the analogy it bears Called "logistical" because of the analogy it bears with the systems of warehouses, depots, and with the systems of warehouses, depots, and distribution channels.distribution channels.

Who Needs Grid Computing?Who Needs Grid Computing?

A chemist may utilize hundreds of processors A chemist may utilize hundreds of processors to screen thousands of compounds per hour.to screen thousands of compounds per hour.

Teams of engineers worldwide pool resources Teams of engineers worldwide pool resources to analyze terabytes of structural data.to analyze terabytes of structural data.

Meteorologists seek to visualize and analyze Meteorologists seek to visualize and analyze data of climate with enormous computational data of climate with enormous computational demands.demands.

Grid UsersGrid Users

Grid developersGrid developers Tool developersTool developers Application developersApplication developers End UsersEnd Users System AdministratorsSystem Administrators

Grid DevelopersGrid Developers

Very small group.Very small group. Implementers of a grid “protocol” who Implementers of a grid “protocol” who

provides the basic services required to provides the basic services required to construct a grid.construct a grid.

Tool DevelopersTool Developers

Implement the programming models used by Implement the programming models used by application developers.application developers.

Implement basic services similar to Implement basic services similar to conventional computing services:conventional computing services: User authentication/authorizationUser authentication/authorization Process managementProcess management Data access and communicationData access and communication

Application DevelopersApplication Developers

Construct grid-enabled applications for end-Construct grid-enabled applications for end-users who should be able to use these users who should be able to use these applications without concern for the applications without concern for the underlying grid.underlying grid.

Provide programming models that are Provide programming models that are appropriate for grid environments and services appropriate for grid environments and services that programmers can rely on when that programmers can rely on when developing (higher-level) applications.developing (higher-level) applications.

System AdministratorsSystem Administrators

Balance local and global concerns.Balance local and global concerns. Manage grid components and infrastructure.Manage grid components and infrastructure. Some tasks still not well delineated due to the Some tasks still not well delineated due to the

high degree of sharing required.high degree of sharing required.

Some Highly-Visible GridsSome Highly-Visible Grids

The NASA Information Power Grid (IPG).The NASA Information Power Grid (IPG). The Distributed Terascale Facility (DTF) The Distributed Terascale Facility (DTF)

Project.Project.

SoftwareSoftware infrastructureinfrastructure

Globus Globus CondorCondor HarnessHarness LegionLegion IBPIBP Net SolveNet Solve

GlobusGlobus

started in 1996 and is gaining popularity year started in 1996 and is gaining popularity year after year.after year.

A project to develop the underlying A project to develop the underlying technologies needed for the construction of technologies needed for the construction of computational grids.computational grids.

Focuses on execution environments for Focuses on execution environments for integrating widely-distributed computational integrating widely-distributed computational platforms, data resources, displays, special platforms, data resources, displays, special instruments and so forth.instruments and so forth.

CondorCondor

The Condor project started in 1988 at the The Condor project started in 1988 at the University of Wisconsin-Madison.University of Wisconsin-Madison.

The main goal is to develop tools to support The main goal is to develop tools to support High Throughput Computing on large High Throughput Computing on large collections of computing resources.collections of computing resources.

LegionLegion

An object-based software project designed at An object-based software project designed at the University of Virginia to support millions the University of Virginia to support millions of hosts and trillions of objects linked together of hosts and trillions of objects linked together with high-speed links. with high-speed links.

Allows groups of users to construct shared Allows groups of users to construct shared virtual work spaces, to collaborate research virtual work spaces, to collaborate research and exchange information.and exchange information.

HarnessHarness

A Heterogeneous Adaptable Reconfigurable A Heterogeneous Adaptable Reconfigurable Networked SystemNetworked System

A collaboration between Oak Ridge National A collaboration between Oak Ridge National Lab, the University of Tennessee, and Emory Lab, the University of Tennessee, and Emory University.University.

IBPIBP

The Internet Backplane Protocol (IBP) is a The Internet Backplane Protocol (IBP) is a middleware for managing and using remote middleware for managing and using remote storage. storage.

It was devised at the University of Tennessee It was devised at the University of Tennessee to support Logistical Networking in large to support Logistical Networking in large scale, distributed systems and applications. scale, distributed systems and applications.

NetSolveNetSolve

A client-server-agent model.A client-server-agent model. Designed for solving complex scientific Designed for solving complex scientific

problems in a loosely-coupled heterogeneous problems in a loosely-coupled heterogeneous environment.environment.

CONCLUSIONCONCLUSION

Grid Computing involves cost savings, speed of Grid Computing involves cost savings, speed of computation, and agility.computation, and agility.

The grid adjusts to accommodate the fluctuating data The grid adjusts to accommodate the fluctuating data volumes that are a typical in the seasonal business.volumes that are a typical in the seasonal business.

Grid Computing takes advantage of the fact that most Grid Computing takes advantage of the fact that most of the computers in United States use their central of the computers in United States use their central processing units on average only 25% of the time for processing units on average only 25% of the time for the work they have been assigned.the work they have been assigned.

THANK YOU….THANK YOU….