• Home
  • Search
  • Is there exploitable thread-level parallelism in general-purpose application programs?
  • Cite Icon1
  • https://doi.org/10.1109/ipdps.2003.1213297Copy DOI Icon

Is there exploitable thread-level parallelism in general-purpose application programs?

  • Apr 22, 2003
  • Pen-Chung Yew
Show More
  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

Summary form only given. Most of the thread-level parallelism (TLP) being successfully exploited so far has been primarily from scientific application programs, in particular, floating-point programs. General-purpose applications, especially those written in C or C++, such as the benchmarks in SPECint2000, have primarily been exploiting only instruction-level parallelism (ILP). A lot of research has been done recently on multiprocessors-on-a-chip (often called "multithreaded processors") because VLSI technology today allows multiple processor cores to be implemented on a single chip. An interesting question has arisen as to how much TLP and ILP could be exploited in general-purpose application programs so such multithreaded processors could become the main work horses of future computer systems. In this paper, we discuss the program characteristics which make it so difficult to exploit TLP in these general-purpose application programs, and present several machine models and simulation techniques for studying how much TLP and ILP could be exploited with these models. We also present some measurements on program characteristics important to the design of such multithreaded processors and their compilers.

Similar Papers
  • Conference Article
  • Citations2

Architectures for one billion of transistors

  • Sep 20, 2000
  • Mateo Valero
  • Research Article
  • Citations42

Multithreaded Processors

  • Mar 01, 2002
  • The Computer Journal
  • T Ungerer
  • Conference Article
  • Citations3

Exploiting Java instruction/thread level parallelism with horizontal multithreading

  • Jan 15, 2001
  • Kenji Watanabe +2
  • Conference Article
  • Citations2

A Multithreaded VLIW Soft Processor Family

  • Apr 01, 2013
  • Kalin Ovtcharov +2
  • Conference Article
  • Citations4

MLP-Aware Dynamic Cache Partitioning

  • Jan 22, 2018
  • Miquel Moretó +3
  • Book Chapter

A Bypass Mechanism to Enhance Branch Predictor for SMT Processors

  • Aug 23, 2007
  • Yongfeng Pan +3
  • Conference Article
  • Citations8

A cycle-approximate, mixed-ISA simulator for the KAHRISMA architecture

  • Mar 12, 2012
  • Timo Stripf +2
  • Conference Article
  • Citations1

Evaluation of Multi-threaded Processor Designs for Energy Efficient Embedded Systems

  • Dec 01, 2013
  • Ran Zhang +1
  • Research Article
  • Citations46

Piranha

  • May 01, 2000
  • ACM SIGARCH Computer Architecture News
  • Luiz André Barroso +8
  • Book Chapter

Simultaneous Multithreading Trace Processors

  • Jan 01, 2003
  • K F Wang +2
  • Research Article
  • Citations19

Performance Evaluation of Dynamic Speculative Multithreading with the Cascadia Architecture

  • Jan 01, 2010
  • IEEE Transactions on Parallel and Distributed Systems
  • David A Zier +1
  • Conference Article
  • Citations19

MLP-Aware Dynamic Cache Partitioning

  • Sep 01, 2007
  • Miquel Moreto +3
  • Conference Article
  • Citations51

A Flexible Heterogeneous Multi-Core Architecture

  • Dec 22, 2017
  • Miquel Pericàs +5
  • Conference Article
  • Citations1

VLIW across multiple superscalar processors on a single chip

  • Nov 11, 1997
  • Soohong P Kim +2
  • Book Chapter

A Compiler Framework for Supporting Speculative Multicore Processors

  • Aug 23, 2007
  • Pen-Chung Yew
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.