Methodology of Computational Science

3/16/00


Click here to start


Table of Contents

Methodology of Computational Science

Abstract of Methodology of Computational Science Presentation

Parallel Computing Methodology in a Nutshell I

Parallel Computing Methodology in a Nutshell II

Potential in a Vacuum Filled Rectangular Box

Basic Sequential Algorithm

Update on the Grid

Parallelism is Straightforward

Communication is Needed

What is Parallel Architecture?

Parallel Computers -- Classic Overview

Distributed Memory Machines

Communication on Distributed Memory Architecture

Distributed Memory Machines -- Notes

Shared-Memory Machines

Communication on Shared Memory Architecture

Shared-Memory Machines -- Notes

Distributed Shared Memory Machines

Summary on Communication etc.

Communication Must be Reduced

Seismic Simulation of Los Angeles Basin

Irregular 2D Simulation -- Flow over an Airfoil

Heterogeneous Problems

Load Balancing Particle Dynamics

Reduce Communication

Minimize Load Imbalance

Parallel Irregular Finite Elements

Irregular Decomposition for Crack

Further Decomposition Strategies

Summary of Parallel Algorithms

Data Parallelism in Algorithms

Functional Parallelism in Algorithms

Pleasingly Parallel Algorithms

Parallel Languages

Data-Parallel Languages

Message-Passing Systems

Shared Memory Programming Model

Structure(Architecture) of Applications - I

Structure(Architecture) of Applications - II

Multi Server Model for metaproblems

Multi-Server Scenario

The 3 Roles of Java

Why is Java Worth Looking at?

What is Java Grande?

Java and Parallelism?

“Pure” Java Model For Parallelism

Pragmatic Computational Science January 2000 I

Pragmatic Computational Science January 2000 II

Email: gcf@npac.syr.edu

Home Page: http://www.npac.syr.edu