• Home
  • Search
  • Characterizing and Mitigating Work Time Inflation in Task Parallel Programs
  • Open Access IconOpen Access
  • Cite Icon11
  • https://doi.org/10.1155/2013/898597Copy DOI Icon

Characterizing and Mitigating Work Time Inflation in Task Parallel Programs

Show More
  • Abstract
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Task parallelism raises the level of abstraction in shared memory parallel programming to simplify the development of complex applications. However, task parallel applications can exhibit poor performance due to thread idleness, scheduling overheads, andwork time inflation– additional time spent by threads in a multithreaded computation beyond the time required to perform the same work in a sequential computation. We identify the contributions of each factor to lost efficiency in various task parallel OpenMP applications and diagnose the causes of work time inflation in those applications. Increased data access latency can cause significant work time inflation in NUMA systems. Our locality framework for task parallel OpenMP programs mitigates this cause of work time inflation. Our extensions to the Qthreads library demonstrate that locality-aware scheduling can improve performance up to 3X compared to the Intel OpenMP task scheduler.

Loading PDF

Similar Papers
  • Conference Article
  • Citations39

Characterizing and mitigating work time inflation in task parallel programs

  • Nov 10, 2012
  • Stephen L Olivier +3
  • Conference Article
  • Citations12

Dynamic Task Scheduling and Load Balancing on Cell Processors

  • Feb 01, 2010
  • Ralf Hoffmann +2
  • Conference Article
  • Citations28

Braid: integrating task and data parallelism

  • Nov 19, 2002
  • E.A West +1
  • Conference Article
  • Citations11

A NUMA-Aware Provably-Efficient Task-Parallel Platform Based on the Work-First Principle

  • Sep 01, 2018
  • Justin Deters +3
  • Conference Article
  • Citations3

A framework for generating task parallel programs

  • Jan 01, 1999
  • U Fissgus +2
  • Conference Article
  • Citations8

Autotuning of a Cut-Off for Task Parallel Programs

  • Sep 01, 2016
  • Shintaro Iwasaki +1
  • Conference Article
  • Citations72

A new model for integrated nested task and data parallel programming

  • Jun 21, 1997
  • Jaspal Subhlok +1
  • Research Article
  • Citations11

A new model for integrated nested task and data parallel programming

  • Jun 21, 1997
  • ACM SIGPLAN Notices
  • Jaspal Subhlok +1
  • Book Chapter
  • Citations4

An integrated annotation and compilation framework for task and data parallel programming in Java

  • Jan 01, 2004
  • Advances in Parallel Computing
  • Henk Sips +1
  • Conference Article
  • Citations4

An Adaptive Task Granularity Based Scheduling for Task-centric Parallelism

  • Aug 01, 2014
  • Jianmin Bi +5
  • PDF
  • Research Article
  • Citations4

Improving the accessibility of NUMA‐aware C++ application development based on the PGASUS framework

  • Feb 16, 2022
  • Concurrency and Computation: Practice and Experience
  • Max Plauth +3
  • Conference Article
  • Citations16

Adding Tightly-Integrated Task Scheduling Acceleration to a RISC-V Multi-core Processor

  • Oct 12, 2019
  • Lucas Morais +6
  • Conference Article
  • Citations7

Atomicity violation checker for task parallel programs

  • Feb 29, 2016
  • Adarsh Yoga +1
  • Research Article

Task Parallel Programming on the Cell Processor

  • Mar 01, 2011
  • it - Information Technology
  • Andreas Prell
  • Research Article
  • Citations32

Models and Scheduling Algorithms for Mixed Data and Task Parallel Programs

  • Dec 01, 1997
  • Journal of Parallel and Distributed Computing
  • Soumen Chakrabarti +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.