• 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.76.1996.tde-16032015-153830
Document
Author
Full name
Wilson Bittencourt Vicentini
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 1996
Supervisor
Committee
Garcia Neto, Alvaro (President)
Ruggiero, Carlos Antonio
Valente, Fredy Joao
Title in Portuguese
Estudo da distribuicao de ocorrencias de instrucoes em grafos de dependencia de dados da arquitetura Wolf
Keywords in Portuguese
Arquiteturas a fluxos de dados
Arquiteturas paralelas
Abstract in Portuguese
As arquiteturas Dataflow apresentam características baseadas em um modelo cujo controle de execução e feito pela disponibilidade dos dados, possibilitando a exploração de paralelismo implícito em um programa. Para contextualizar o trabalho, são apresentadas as principais implementações em fluxo de dados. Este trabalho estuda a distribuição da ocorrência de instruções em Grafos de Dependência de Dados da arquitetura Wolf desenvolvida no IFSC. O estudo realizado concluiu que existe urna grande concentração na utilização das instruções por parte dessa arquitetura: das cento e uma instruções da maquina, cinco apresentam, em media, de trinta por cento a cinqüenta por cento do total. Essa popularidade de instruções ocorre em todas as formas de analise e para todos os benckmarks investigados. Os resultados desse trabalho fornecem uma base para a implementação mais eficiente dessas arquiteturas, e indicam que qualquer mecanismo que explore essa popularidade, produzirá ganho
Title in English
Not available
Keywords in English
Not available
Abstract in English
The Dataflow architecture presents characteristics based on a model in which execution control is done by the data availability, making the exploitation of the implicit parallelism possible in a program. The main implementation in data flux is showed, having in view to put it inside the work's context. This work studies the distribution of the occurrence of the instructions in Graphs of Data Dependence of the Wolf architecture developed at IFSC. The present study concluded that there is a large concentration on the utilizations of the instructions in this architecture: from the one hundred and one instructions of the machine, five present, in average, from thirty to fifty per cent of the total. This popularity of instructions occurs in all forms of analysis and for all investigated benckmarks. The results of this work furnish a base for a more efficient implementation of these architectures and they indicate that any mechanism, which explores this popularity, will produce gains
 
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
2015-03-19
 
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.