• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.45.2015.tde-20230727-113135
Document
Author
Full name
Ademar Marques Lacerda Filho
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2015
Supervisor
Title in Portuguese
IntLAG: uma biblioteca para álgebra linear intervalar em processadores gráficos
Keywords in Portuguese
Ciencia Da Computacao
Abstract in Portuguese
A aritmética intervalar é utilizada em problemas físicos e computacionais onde uma solução rigorosa é necessária. Muitos destes problemas são computacionalmente caros e requerem implementações de alto desempenho para que a solução seja viável. Este trabalho visa a implementação de uma biblioteca de código aberto do tipo BLAS para aritmética intervalar utilizando conceitos de GPGPU por meio da extensão CUDA e a comparação desta com técnicas mais tradicionais. Apresentamos a biblioteca e sua abordagem focando no aspecto de implementação da mesma. Por fim, apresentamos um benchmark comparando o desempenho dos métodos desenvolvidos em GPU com suas contrapartes em CPU, seriais e paralelas, resultados promissores são apresentados em uma placa gráfica de última geração.
Title in English
not available
Abstract in English
Interval arithmetic is used by physical and computational scenarios in which a rigorous solution is required. Many of these problems are computationally expensive and require high-performance implementations. This work aims to implement an open source BLAS type library for interval arithmetic using GPGPU concepts through CUDA extension and to compare it with more traditional techniques. We present the library and its approach focusing on the implementation aspects. Finally, we present a benchmark comparing the performance of the methods developed in GPU with their counterparts in CPU, serial and parallel both, promissing results are presented in a high-end GPU.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2023-07-27
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.