Towards a parallel component in a GPU-CUDA environment: a case study with the L-BFGS Harwell routine

Modern graphics processing units (GPUs) have been at the leading edge of increasing parallelism over the last 10 years. This fact has encouraged the use of GPUs in a broader range of applications, where developers are required to lever age this technology with new programming models which ease the t...

Full description

Saved in:
Bibliographic Details
Published in:International journal of computer mathematics Vol. 92; no. 1; pp. 59 - 76
Main Authors: D'Amore, L., Laccetti, G., Romano, D., Scotti, G., Murli, A.
Format: Journal Article
Language:English
Published: Abingdon Taylor & Francis 02.01.2015
Taylor & Francis Ltd
Subjects:
Algorithms
Central processing units
Codes
Computer programming
Computer programs
CPUs
Data assimilation
Developers
GPU computing
Graphics processing units
Guidelines
L-BFGS routine
Mathematical models
mathematical software
Optimization algorithms
parallel computing
Parallel processing
Routines
Software engineering
ISSN:0020-7160, 1029-0265
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Modern graphics processing units (GPUs) have been at the leading edge of increasing parallelism over the last 10 years. This fact has encouraged the use of GPUs in a broader range of applications, where developers are required to lever age this technology with new programming models which ease the task of writing programs to run efficiently on GPUs. In this paper, we discuss the main guidelines to assist the developer when porting sequential scientific code on modern GPUs. These guidelines were carried out by porting the L-BFGS , the (Limited memory-) BFGS algorithm for large scale optimization, available as Harwell routine VA15. The specific interest in the L-BFGS algorithm arises from the fact that this is the computational module with the longest running time of a Oceanographic Data Assimilation application software, on which some of the authors are working.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0020-7160
1029-0265
DOI:10.1080/00207160.2014.899589