Automatic Coarse Grain Task Parallel Processing
Using OSCAR Multigrain Parallelizing Compiler
Hironori Kasahara, Motoki Obata, Kazuhisa Ishizaka
- Full Paper(.ps version)
This paper describes an efficient implementation method
of automatic coarse grain task parallel processing using OSCAR
multigrain parallelizing compiler for a SMP machine.
The coarse grain task parallel processing is important to improve
the effective performance of wide range of multiprocessor systems
from a single chip multiprocessor to a high performance computer
beyond the limit of the loop parallelism.
The performance evaluation of the proposed method shows that OSCAR
compiler gives us 6.3 times speed up in average for five application
programs from SPEC 95fp and Perfect Benchmarks for which IBM XL
Fortran compiler gives us 3.3 times speed up in average on RS6000
SP 604e High Node 8-processor SMP machine.
Please contact our
webadmin with any comments or changes.
Unless explicitly stated otherwise, all material is
copyright © The University of Edinburgh.