The Shared Memory Parallelisation of an Ocean Modelling Code Using an Interactive Parallelisation Toolkit

Authors:
Cos S. Ierotheou;S. P. Johnson;P. F. Leggett;Mark Cross
Affiliations:
-;-;-;-
Venue:
ICCS '02 Proceedings of the International Conference on Computational Science-Part II
Year:
2002

Citing 9
Cited 0

Computer aided parallelisation tools (CAPTools)—conceptual overview and performance on the parallelisation of structured mesh codes

Parallel Computing
Exploitation of symbolic information in interprocedural dependence analysis

Parallel Computing
Automatic parallel code generation for message passing on distributed memory systems

Parallel Computing
Integrating user knowledge with information from parallelisation tools to facilitate the automatic generation of efficient parallel FORTRAN code

Parallel Computing
An ocean model code for array processor computers

Computers & Geosciences
Automatic code generation of overlapped communications in a parallelisation tool

Parallel Computing
Automatic and effective multi-dimensional parallelisation of structured mesh based codes

Parallel Computing
Automatic Generation of OpenMP Directives and Its Application to Computational Fluid Dynamics Codes

ISHPC '00 Proceedings of the Third International Symposium on High Performance Computing
Validity of the single processor approach to achieving large scale computing capabilities

AFIPS '67 (Spring) Proceedings of the April 18-20, 1967, spring joint computer conference

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper briefly describes an interactive parallelisation toolkit that can be used to generate parallel code suitable for either a distribtued memory system (using message passing) or a shared memory system (using OpeitMP). This study focuses on how tile toolkit is used to parallelise a complex heterogeneous ocean modelling code within a few horns for use on a shared memory parallel system. The generated parallel code is essentially tile serial code with OpeitMP directives added to express tile parallelism. The results show that substantial gains in perfomlance can be achieved over the single ttu-ead version with very little effort.