vigoni:fut_rmi:book:05 (In a collection)
|
Author(s) | Marco Aldinucci, Marco Danelutto, Jan Dünnweber and Sergei Gorlatch |
Title | « Optimization techniques for skeletons on grids » |
In | Grid Computing and New Frontiers of High Performance Processing |
Series | Advances in Parallel Computing |
Editor(s) | L. Grandinetti |
Volume | 14 |
Page(s) | 255-273 |
Year | 2005 |
Publisher | Elsevier |
ISBN number | 0-444-51999-8 |
ISSN number | 09275452 |
URL | http://calvados.di.unipi.it/storage/paper_files/2005_LithiumFutRMI_book.pdf |
Abstract |
Skeletons are common patterns of parallelism, such as farm and pipeline, that can be abstracted and offered to the application programmer as programming primitives. We describe the use and implementation of skeletons on emerging computational grids, with the skeleton system Lithium, based on Java and RMI, as our reference programming syttem. Our main contribution is the exploration of optimization techniques for implementing skeletons on grids based on an optimized, future-based RMI mechanism, which we integrate into the macro-dataflow evaluation mechanism of Lithium. We discuss three optimizations: 1) a lookahead mechanism that allows to process multiple tasks concurrently at each grid server and thereby increases the overall degree of parallelism, 2) a lazy taskbinding technique that reduces interactions between grid servers and the task dispatcher, and 3) dynamic improvements that optimize the collecting of results and the work-load balancing. We report experimental results that demonstrate the improvements due to our optimizations on various testbeds, including a heterogeneous grid-like environment. |
@incollection{vigoni:fut_rmi:book:05,
volume = {14},
month = oct,
chapter = {2},
issn = {09275452},
author = {Marco Aldinucci and Marco Danelutto and Jan D{\"u}nnweber and Sergei
Gorlatch},
series = {Advances in Parallel Computing},
booktitle = {Grid Computing and New Frontiers of High Performance Processing},
editor = {L. Grandinetti},
url = {http://calvados.di.unipi.it/storage/paper_files/2005_LithiumFutRMI_book.pdf},
title = {Optimization techniques for skeletons on grids},
abstract = {Skeletons are common patterns of parallelism, such as farm and
pipeline, that can be abstracted and offered to the application
programmer as programming primitives. We describe the use and
implementation of skeletons on emerging computational grids, with
the skeleton system Lithium, based on Java and RMI, as our
reference programming syttem. Our main contribution is the
exploration of optimization techniques for implementing skeletons
on grids based on an optimized, future-based RMI mechanism, which
we integrate into the macro-dataflow evaluation mechanism of
Lithium. We discuss three optimizations: 1) a lookahead mechanism
that allows to process multiple tasks concurrently at each grid
server and thereby increases the overall degree of parallelism, 2)
a lazy taskbinding technique that reduces interactions between
grid servers and the task dispatcher, and 3) dynamic improvements
that optimize the collecting of results and the work-load
balancing. We report experimental results that demonstrate the
improvements due to our optimizations on various testbeds,
including a heterogeneous grid-like environment.},
isbn = {0-444-51999-8},
publisher = {Elsevier},
year = {2005},
pages = {255-273},
}
This document was generated by bib2html 3.3.
(Modified by Luca Paolini, under the GNU General Public License)