Algorithmic-parameter optimization of a parallelized split-step Fourier transform using a modified BSP cost model

Elankovan A Sundararajan, Malin Premaratne, Shanika Karunasekera, Aaron Harwood

Research output: Contribution to journalArticle

Abstract

Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.

Original languageEnglish
Pages (from-to)233-244
Number of pages12
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3358
Publication statusPublished - 2004
Externally publishedYes

Fingerprint

Cost Model
Parameter Optimization
Fourier Analysis
Fourier transform
Fourier transforms
Costs and Cost Analysis
Costs
Adaptive algorithms
Adaptive Algorithm
Parallel Processing
Parallel algorithms
Computational Model
Parallel Algorithms
Model-based
Predict
Resources
Processing
Model
Research

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

@article{f19fa1a745724f63a53987618ec76a4b,
title = "Algorithmic-parameter optimization of a parallelized split-step Fourier transform using a modified BSP cost model",
abstract = "Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.",
author = "{A Sundararajan}, Elankovan and Malin Premaratne and Shanika Karunasekera and Aaron Harwood",
year = "2004",
language = "English",
volume = "3358",
pages = "233--244",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Algorithmic-parameter optimization of a parallelized split-step Fourier transform using a modified BSP cost model

AU - A Sundararajan, Elankovan

AU - Premaratne, Malin

AU - Karunasekera, Shanika

AU - Harwood, Aaron

PY - 2004

Y1 - 2004

N2 - Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.

AB - Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.

UR - http://www.scopus.com/inward/record.url?scp=35048886495&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048886495&partnerID=8YFLogxK

M3 - Article

VL - 3358

SP - 233

EP - 244

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -