We briefly discuss why GEM is interesting |
We state the computational problem for GEM and analyze parallel computing issues for key steps |
We map into a particle dynamics analogy using Green's function formalism and consider O(N2) cut-off O(N) and fast multipole methods |
We go in detail through the fast multipole method |
We expect object oriented approaches to be very important booth at language level and in using technologies like CORBA to integrate world-wide collaboration, simulations and experiments. |
001 Computational Science and N Body algorithms Illustrated by GEM: General Earthquake Simulation Project CPS615 Introduction to Computational Science October 98 002 Abstract of GEM Analysis for Computational Science 003 Earthquakes are Worldwide 004 Northridge Earthquake 1994 (Southern California) 005 Southern California Earthquake Activity 006 We need to predict earthquakes! 007 Possible Special Features of Earthquake Simulation 008 Basic Computational Structure - I 009 Basic Computational Structure - II 010 Analysis of Computational Structure 011 First two Solutions of O(N2) Computational Complexity 012 Second two Solutions of O(N2) Computational Complexity 013 Basic Idea of Fast Multipole Algorithm 014 Intermediate results of a computation of 322 million particles on ASCI Red 015 Intermediate results of a computation of 9.7 million particles on PC Cluster loki 016 Some Performance Results of Interest from Salmon and Warren 017 Hierarchical Breakup of 2D Space 018 Simple Illustration of Tree Data Structure 019 Tree Structure for 10,000 bodies centrally clustered in a disk 020 Generation of Tree for a small number of particles 021 3 Approximations to Force on a Particle in Fast Multipole Approach 022 Parallelism in O(N2) N Body Approach I 023 Parallelism in O(N2) N Body Approach II 024 Parallelism in Cut Off Force Approach 025 Problems in Cut off Force Parallelism 026 Cyclic and Block Decomposition for Graphics Ray Tracing 027 Generation of Keys in Salmon Warren Method 028 Generation of 3D Key for Salmon Warren 029 Parallelism in Salmon Warren Approach 030 Two Space Filling Curves 031 Morton Curve split up into 8 processors represented by different gray levels 032 Space Filling Curve chopped up into equal length parts 033 Parallel Algorithm in Fast Multipole I 034 Locally essential Data for Processor in Bottom Left Hand Corner of Processor Array 035 Parallel Algorithm in Fast Multipole II 036 Scaling Ideas in GEM 037 Different Physical Scales in GEM 038 Lessons from Other Fields