Lazy evaluation of PDE coefficients in the EScript system

Authors:
Joel Fenwick;Lutz Gross
Affiliations:
The University of Queensland, St Lucia, QLD, Australia;The University of Queensland, St Lucia, QLD, Australia
Venue:
AusPDC '10 Proceedings of the Eighth Australasian Symposium on Parallel and Distributed Computing - Volume 107
Year:
2010

Citing 7
Cited 0

Conception, evolution, and application of functional programming languages

ACM Computing Surveys (CSUR)
The Haskell: The Craft of Functional Programming

The Haskell: The Craft of Functional Programming
Modern Compiler Implementation in Java

Modern Compiler Implementation in Java
UNIX for Programmers and Users (3rd Edition)

UNIX for Programmers and Users (3rd Edition)
Essentials of Programming Languages, 3rd Edition

Essentials of Programming Languages, 3rd Edition
Programming Language: Principles and Paradigms

Programming Language: Principles and Paradigms
A python module for PDE-Based numerical modelling

PARA'06 Proceedings of the 8th international conference on Applied parallel computing: state of the art in scientific computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

EScript is an extension to Python for solving partial differential equations on parallel computers. It is parallelised for both MPI and shared memory, multi-core systems using OpenMP. In this paper, we discuss lazy evaluation as a strategy to reduce the cost of evaluating the coefficients of PDEs prior to solving. We show that our implementation provides significant memory and time savings for a problem involving complex expressions.