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
Complex Systems and Parallel Computing Australian International Conference on Complex Systems Australian National University Canberra, Australia December 14-16, 1992 Geoffrey C. Fox 2
The Three Themes of Lecture: Parallel Computers and Complex Systems 3
Issues in Parallel Computers for the Simulation of Complex Systems 4
Standard Performance Graph Heading to 1 to 10 Teraflops by year 2000 5
When will parallel computing take over? 6
The President's High Performance Computing and Communication Initiative (HPCCI) 7
Challenges and Status of Parallel Computing 8
High Performance Fortran Overview 9
HPF computational model 10
Example of Fortran-90D source code: Gaussian Elimination 11
HPF directives 12
Data Alignment and Distribution Directives 13
Examples of Alignments (1) 14
Examples of Distributions (1) 15
For More Information on HPF 16
FORTRAN-90D The First Implementation of HPF (NPAC, Syracuse University) Current Status 17
Common Software needed for Heterogeneous Local Area Network (Ethernet - FIDDI - HIPPI - FCS ......) 18
Importance of MetaProblems 19
Hybrid Problem Structure for Command and Control 20
The Mapping of Heterogeneous Problems onto Heterogeneous Computer Systems 21
SIMCITY is an interesting PC based complex system simulation. 22
Implementation of Complex System Simulation 23
AVS as System Integration Tool 24
Parallel AVS - Planned Project at NPAC 25
Architecture of Parallel AVS System 26
VR Operating Shells 27
Components of Proposed Televirtuality Server at NPAC 28
A Theory of Parallel Computing based on Complex Systems 29
Computing as a set of Mapping Problems 30
Complex Systems to give a theory of computing 31
Parallel Computing is "just" an optimization problem, even if we can't agree on what to optimize 32
Concurrent Computation as a Mapping Problem -I 33
Concurrent Computation as a Mapping Problem - II 34
Computation as a map of a set of Complex Systems 35
Domain Decomposition and Complex Systems ? 36
Physical Analogy for Complex Computer 37
The Physical Space/TimeAnalogy for a General Problem 38
Some Temporal Properties of Computation 39
General Space Time Complex System Picture for Problem to Computer Mapping 40
Computer Languages and Space - Time Properties 41
Information Dimension of a General Complex System 42
Performance of a Parallel Computer 43
Hierarchical Multicomputer Spatial and Temporal Decomposition 44
Shared or Hierarchical Memory Computer 45
Comparison of Cache and Distributed Memory Communication Overhead 46
Extension of Space-Time Picture to treat Hierarchial memory and caches etc. 47
Space-Time Decompositions for the concurrent one dimensional wave equation 48
Typical Example of Mapping an Irregular Bunch of Grid Points 49
Use of Physical Optimization in High Performance Fortran 50
Physics Analogy for Load Balancing 51
Complex System SHLSoft governed by Hamiltonian = Execution Time 52
Decomposition of an Arch onto 16 Processors in a Hypercube 53
PHYSICS ANALOGY FOR STATIC AND DYNAMIC LOAD BALANCING 54
General definition of temperature TS of a complex system 55
Particle dynamics problem on a four node system 56
Instantaneous Energy Distribution for Time Dependent Domain Decomposition and Block Scattered Distributions 57
Time Averaged Energy for Adaptive Particle Dynamics Problem 58
A general theory of computation 59
HISTORICALLY ONE OF THE MOTIVATIONS FOR THE RESEARCH WAS TO " AUTOMATE" THE KNOWN FOLD ALGORITHM 60
The String Formalism for Dynamic Computations 61
Loosely Synchronous Static and Adaptive Problems in the String Picture 62
An initial approach to computational string dynamics or equivalently the Construction of the Energy Function 63
Full String Dynamics as an Interacting Field Theory 64
Complex systems suggest new computational methodologies 65
Physical Optimization and Computation Approaches and their Field of Origin 66
Genetic Algorithms for Data Decomposition 67
Three Major Genetic Operators 68
MultiScale Methods in Parallel Data Decomposition 69
Results of Various Physical Optimization Methods for Data Decomposition 70
A similar but Larger Problem 71
Some Overall Questions Relevant In Classisfying Optimization Problems and Methods 72
Two Types of Global Mininum and their relation to Local Minima 73
Typical Formalism for Physical Optimization 74
Global and Local Minima in Temperature Dependent Free Energy 75
Comparison of Physical Optimization Methods 76
Some Applications of Deterministic Annealing 77
Simulated Tempering -- a New Approach to Monte Carlo Optimization/Simulated Annealing 78
The Conventional Simulated Annealing and its Problems for Random Field Ising Models 79
Key Idea in The Tempering Approach 80
Goodbye! Many Choices - Which is best When? 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.