Foilset Search Full Index for Basic foilset

Overview of HPC for Physical Sciences

Given by Geoffrey C. Fox at Mardi Gras Conference on Concurrent Computing in Physical Sciences on February 18, 1993. Foils prepared October 22 1997

This overviews status of HPCC architectures
Software approaches focussing on HPF and an Application Analysis
Problem Architectures Load Balancing and the Complex System Approach

Table of Contents for Overview of HPC for Physical Sciences

There are two types of foils -- html and image which are each available in basic and JavaScript enabled "focused" style
(basic:)(focus style:) Denote Foils where Image Critical
(basic:)(focus style:) Denote Foils where HTML is sufficient
1 Overview of High Performance Computing for Physical Sciences
2 Performance of Computers as a Function of Time
3 Performance of QCD Machines
4 When will Parallel Computing Take Over ?
5 Parallelism Implies
6 Program in Computational Science Implemented within current academic framework
7 We have learnt that Parallel Computing Works !
8 METHODOLOGY
9 Sideways View of Concurrent Computing Mappings
10 Sample uses of Concurrent Computers
- 4 Nodes
11 Parallel Computing Trends to Year 2000
12 39 National Grand Challenges
13 39 National Grand Challenges (II)
14 39 National Grand Challenges (III)
15 39 National Grand Challenges (IV)
16 Most Grand Challenges are
17 ACTION Motivation
18 Applications of Interest to Industry IA
19 Applications of Interest to Industry IB
20 Applications of Interest to Industry IC
21 Pluses and Minuses of Parallel Computing
22 P C E T E C H Ñ A Result of
23 The Life History of an HPC Idea
24 Core Enabling Software Technologies
25 Core Enabling Algorithms and Components
26 Portable Scalable Languages
27 Problem Architectures
28 Software Problem Architecture Interactions
29 Why build on existing languages - especially Fortran / C
30 Fortran( C ) Plus Message Passing
31 Fortran( C ) Plus Message Passing
32 Data Parallel Fortran
33 Data Parallel Fortran ( C, C++, LISP, ADA )
34 The Origins of High Performance Fortran
35 Strategy for HPF
36 Model for Scalable Languages
37 Dynamically Triangulated Random Surfaces
38 Update Strategies for DTRS
39 N=36 Wireframe DTRS
40 N=144 Lambda =1.5 Colored DTRS
41 N=144 Lambda=1 Colored DTRS
42 Computational Issues and Code Performance Optimization
43 Sequential Computer Architecture
44 Performance of Random Surface Code
45 Parallel Algorithms
46 Parallel Grid Generation
47 What is Fortran 90?
48 HPF Language Features - 1
49 HPF Language Features - 2
50 FORTRAN-90D
The First Implementation of HPF
(NPAC, Syracuse University)
Current Status
51 Fortran90D Performance on Gaussian Elimination
52 Gaussian Elimination
Example of Fortran90D Source Code
53 Fortran90D Interface
54 HPF Directives
55 Data Alignment
56 Data Distribution
57 FORALL
58 DO INDEPENDENT
59 HPF Library
60 Intrinsic Library
61 HPF Library
62 Fortran 90 Local Routine Intrinsics
63 Does HPF need a different type of compiler ?
64 HPF Interpreter
65 Benchmarking Set ( Fortran90D and Fortran 77 )
66 HPF/FORTRAN-90D Benchmark Suite
67 The final draft of the HPF Language Specification is version 1.0 - DRAFT, dated January 25, 1993
68 Request for Public Comment on High Performance Fortran
69 Application Structure Versus Fortran90D Features
70 What applications does HPF support?
71 Current HPF can do I .....
72 Current HPF can do II.....
73 HPF can express using FORALL
74 Classic Irregular Mesh Application
75 HPF Can Express
76 Simulation of Spin Models of Magnets
77 New Monte Carlo Algorithms
78 Magnetization for Random Field Ising Model -- Metropolis
79 Magnetization for Random Field Ising Model -- Simulated Annealing
80 Wolff cluster (bands shown in yellow) for 3 state Potts model at Tc
81 Swendsen-Wang clusters (boundaries shown in black) for 3 state Potts model at Tc
82 Cluster Algorithm versus Metropolis
83 Parallel Algorithms
84 Parallel Cluster Algorithms
85 MIMD algorithm on nCUBE-1
86 SIMD algorithms on CM-2
87 Autocorrelation Plots for Cluster Algorithms
88 HPF probably cannot express well
89 I do not know how to express - let alone optimize -
90 Late breaking results
91 Large N-Body Calculations (Quinn, Salmon, Warren)
92 Hierarchical Decomposition
93 Ncube Speed Up of Barnes Hut Algorithm
94 17x106 Body Simulation
Diameter 250Mpc
Quinn, Salmon, Warren, Zurek
95 The largest "galaxy" halo (137,000 bodies) from the 8.8M body simulation
(Warren, Fullagar, Quinn, Zurek)
96 8M bodies - 10 Mpc diameter
Final state with ~700 resolved "galaxies"
(Warren, Quinn, Zurek)
97 Smooth Particle Hydro simulation of the collision of two "main sequence stars." 137,000 bodies (Warren,Davies)
98 Expressing Problem Structure in Language
99 The map of Problem ---> Computer is performed in two or more stages
100 From Problem to Machine Space as a Mapping
101 Different Granularities of Decomposition I
102 Different Granularities of Decomposition II
103 Compiler/Interpreter Tradeoffs at Different Levels of Granularity
104 The Mapping of Heterogeneous MetaProblems onto Heterogeneous MetaComputer Systems
105 Mapping of Irregular Grid
106 Finding General Maps for FortranD/HPF
107 Three Physical Optimization Methods for Allocating Data to Multicomputer Nodes
108 Finite Element Mesh
109 Graph Contraction
110 Some Results of Load Balancing Studies I
111 Some Results of Load Balancing Studies II
112 Physics Analogy for Load Balancing
113 Complex System SHLSoft governed by Hamiltonian = Execution Time
114 PHYSICS ANALOGY FOR STATIC AND DYNAMIC LOAD BALANCING
115 Definition of Temperature for a Complex System
116 Particle dynamics problem on a four node system
117 Energy Structure in Physics Analogy with Multiple Minima
118 Phase Transitions in Physical model -- Scattered versus Domain Decompositions
Full WebWisdom URL and this Foilset Search
This contains all WebWisdom links preceded by those referenced in this foilset
© Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu

If you have any comments about this server, send e-mail to webmaster@npac.syr.edu.

Page produced by wwwfoil on Wed Oct 22 1997