LOCAL foilset Master Foilset for MPI HPF and DAGH for Parallelization of Black Hole Codes

Given by Scott Klasky, Tom Haupt, Geoffrey C. Fox at SIAM Workshop Minneapolis "Is Message Passing Obsolete" on 16 March 97. Foils prepared 14 March 97
Abstract * Foil Index for this file

We describe basic physics and computational features of Binary Black Collision Grand Challenge

We show a simple finite difference is complicated by Adaptive Mesh refinement Complex inner and outer boundary conditions Need for several algorithms for hyberbolic and elliptic parts of problem

MPI or more generally message passing is not obsolete but rather the only complete solution!

Higher level systems DAGH and especially HPF cannot express full complexity of problem and MPI must fill in the missing parts

HPF2 might be able to express full problem but HPF1 misses many key capabilities

DAGH is more succesful than HPF1 but needs MPI for parts of problem

Postscript or FramemakerVersion

HPF Code for Linear Waves

Grand Challenge Alliance Home Page

NPAC Activity in Grand Challenge Alliance

Table of Contents for full HTML of Master Foilset for MPI HPF and DAGH for Parallelization of Black Hole Codes

This table of Contents Abstract

Foil 1 HPF & DAGH usage in the Binary Black Hole Grand Challenge

Foil 2 Goals of the Alliance

Foil 3 Computational Infrastructure

Foil 4 ADM Equations

Foil 5 ADM Equations continued

Foil 6 Causal Differencing

Foil 7 Causal Differencing continued

Foil 8 Apparent Horizon boundary conditions

Foil 9 Difficulties of the BBH Problem

Foil 10 DAGH System

Foil 11 Parallel Execution in DAGH

Foil 12 Parallel Execution in DAGH continued

Foil 13 Implementation

Foil 14 Sample of HPF and DAGH code

Foil 15 Linear Waves

Foil 16 Black holes (Inner boundary condition)

Foil 17 Black holes continued

Foil 18 Linear wave problem with outer boundary condition

Foil 19 Moving Inner boundary

Foil 20 Moving Inner boundary continued

Foil 21 Finding the Apparent Horizon

Foil 22 Elliptic Problem for the shift

Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu