Understanding Computational Grids: Motivation, Definition, and Characteristics - Prof. Ala, Study notes of Computer Science

An overview of computational grids, explaining the motivation behind their development, defining what they are, and discussing their characteristics. It covers various use cases, such as biochemistry, physics, civil engineering, and climate science, and compares grids to traditional distributed computing systems. The document also includes a checklist for identifying grid characteristics and examples of grids in the wild, such as nasa's information power grid and nsf's teragrid.

Typology: Study notes

Pre 2010

Uploaded on 02/13/2009

koofers-user-p0f
koofers-user-p0f 🇺🇸

10 documents

1 / 4

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
1
Computational
Grids
Derived from:
“The Anatomy of the Grid”, Foster, Kesselman and Tuecke
“Grids: The Top Ten Questions”, Schopf, Nitzberg
“What is The Grid ?”, Foster
“Computational Grids”, Foster and Kesselman
Nitin Madnani, [email protected]
2
Outline
Outline
Motivation
Defining a Grid
Is <X> a grid ?
Grids in the Wild
Last Six Years
Take Home Lessons
References
3
Motivation
Motivation
Group of participants
Varying degrees of prior relationship
Mutually Distrustful
Need to share resources to perform a task
Direct access to software, data, sensors and
computers
Sharing subject to a set of constraints
What, Who, When etc.
Dynamic sharing relationships
4
Motivation (
Motivation (contd
contd)
)
Biochemists exploit 10,000 computers to screen 100,000
compounds in an hour
1,000 physicists worldwide pool resources for petaop
analyses of petabytes of data
Civil engineers collaborate to design, execute & analyze
shake-table experiments
Climate scientists visualize, annotate, & analyze terabyte
simulation datasets
An emergency response team couples real time data,
weather model, population data
5
Motivation (
Motivation (contd
contd)
)
Regular distributed computing
Cannot coordinate across multiple sites
Cannot accommodate multiple resource types
Offers limited form of sharing
Needs centralization
Not good enough
6
Enter the Grid
Enter the Grid
Introduced in mid 90s
Distributed computing for “Big Science”
Hardware/Software Infrastructure
Pools resources to provide computing that is:
Dependable
Consistent
Inexpensive
Pervasive
pf3
pf4

Partial preview of the text

Download Understanding Computational Grids: Motivation, Definition, and Characteristics - Prof. Ala and more Study notes Computer Science in PDF only on Docsity!

Computational

Grids

Derived from: “The Anatomy of the Grid”, Foster, Kesselman and Tuecke “Grids: The Top Ten Questions”, Schopf, Nitzberg “What is The Grid ?”, Foster “Computational Grids”, Foster and Kesselman

Nitin Madnani, [email protected] 2

OutlineOutline

„ Motivation

„ Defining a Grid

„ Is a grid?

„ Grids in the Wild

„ Last Six Years

„ Take Home Lessons

„ References

3

MotivationMotivation

„ Group of participants

„ Varying degrees of prior relationship

„ Mutually Distrustful

„ Need to share resources to perform a task

Direct access to software, data, sensors and

computers

Sharing subject to a set of constraints

„ What, Who, When etc.

Dynamic sharing relationships

4

Motivation (Motivation (contdcontd))

„ Biochemists exploit 10,000 computers to screen 100,

compounds in an hour

„ 1,000 physicists worldwide pool resources for petaop

analyses of petabytes of data

„ Civil engineers collaborate to design, execute & analyze

shake-table experiments

„ Climate scientists visualize, annotate, & analyze terabyte

simulation datasets

„ An emergency response team couples real time data,

weather model, population data

5

Motivation (Motivation (contdcontd))

„ Regular distributed computing

Cannot coordinate across multiple sites

Cannot accommodate multiple resource types

Offers limited form of sharing

Needs centralization

„ Not good enough

6

Enter the GridEnter the Grid

„ Introduced in mid 90s

„ Distributed computing for “Big Science”

„ Hardware/Software Infrastructure

„ Pools resources to provide computing that is:

Dependable

Consistent

Inexpensive

Pervasive

7

What exactly is a grid ?What exactly is a grid?

„ Funding Concept/Marketing Slogan

„ Need for a clear definition

The red-blue cluster is NOT a grid.

„ Architecture is not as important

„ Is there a checklist?

8

The Grid ChecklistThe Grid Checklist

„ Collaboration

Computation- and data-rich environments

„ Coordination of heterogeneous resources across domains

No local management

„ Standard/Open Interfaces & Protocols

Not application specific

„ Guarantees Quality of Service (QoS)

Response Time, Throughput, Availability

„ Transparent to users

9

Virtual OrganizationsVirtual Organizations

„ Set of individuals defined by sharing rules

„ Dimensions

Number & type of participants

Resources being shared

Types of activities

„ Examples:

Storage Service Providers

SETI@HOME participants

Members of a High-Energy Physics Consortium

10

Illustrative ExampleIllustrative Example

11

Is a grid ?Is a grid?

„ X = Cluster Management Systems

QoS Guarantees [ Yes ]

Distributed Resources [ Yes ]

De-centralized Coordination [ No ]

Standard/Open Protocols [ No ]

„ Verdict: Not a grid. A resource.

„ Examples

Sun Grid Engine

Veridian Portable Batch System

12

Is a grid? (Is a grid? (contdcontd))

„ X = The Web

QoS Guarantees [ No ]

Distributed Resources [ Yes ]

De-centralized Coordination [ No ]

Standard/Open Protocols [ Yes ]

„ Verdict: Not a grid.

19

ReferencesReferences

„ I. Foster, C. Kesselman, S. Tuecke, “The Anatomy of the Grid: Enabling Scalable Virtual Organizations” , International J. Supercomputer Applications, 15(3), 2001

„ I. Foster, “What is the Grid? A Three Point Checklist” , GRIDToday, July 20, 2002.

„ J.M. Schopf and B. Nitzberg, “Grids: Top Ten Questions ”, Scientific Programming, 10(2):103 - 111, August 2002.

„ http://www.globus.org/alliance/publications/papers.php