• Cite Icon16
  • https://doi.org/10.1145/263764.263771Copy DOI Icon

Dynamic pointer alignment

  • Jun 21, 1997
  • Xingbin Zhang +1 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Loop tiling and communication optimization, such as message pipelining and aggregation, can achieve optimized and robust memory performance by proactively managing storage and data movement. In this paper, we generalize these techniques to pointer-based data structures (PBDSs). Our approach, dynamic pointer alignment (DPA), has two components. The compiler decomposes a program into non-blocking threads that operate on specific pointers and labels thread creation sites with their corresponding pointers. At runtime, an explicit mapping from pointers to dependent threads is updated at thread creation and is used to dynamically schedule both threads and communication, such that threads using the same objects execute together, communication overlaps with local work, and messages are aggregated. We have implemented DPA to optimize remote reads to global PBDSs on parallel machines. Our empirical results on the force computation phases of two applications that use sophisticated PBDSs, Barnes-Hut and FMM, show that DPA achieves good absolute performance and speedups by enabling tiling and communication optimization on the CRAY T3D.

Similar Papers
  • Research Article
  • Citations1

Dynamic pointer alignment

  • Jun 21, 1997
  • ACM SIGPLAN Notices
  • Xingbin Zhang +1
  • Research Article
  • Citations4

Architectural Support and Mechanisms for Object Caching in Dynamic Multithreaded Computations

  • Aug 01, 1999
  • Journal of Parallel and Distributed Computing
  • Vijay Karamcheti +1
  • Conference Article
  • Citations18

A Hierarchical Load-Balancing Framework for Dynamic Multithreaded Computations

  • Nov 07, 1998
  • Vijay Karamcheti +1
  • Conference Article
  • Citations17

A Hierarchical Load-Balancing Framework for Dynamic Multithreaded Computations

  • Jan 01, 1998
  • V Karamcheti +1
  • Conference Article
  • Citations27

Analysis of TCP and DTN Retransmission Algorithms in Presence of Channel Disruptions

  • Jul 20, 2009
  • Carlo Caini +4
  • Book Chapter
  • Citations1

Scaling FMM with Data-Driven OpenMP Tasks on Multicore Architectures

  • Jan 01, 2016
  • Abdelhalim Amer +6
  • Conference Article
  • Citations2

Parallelizing irregular algorithms

  • Oct 21, 2011
  • Pedro Monteiro +2
  • Research Article
  • Citations4

Structure-driven optimizations for amorphous data-parallel programs

  • Jan 09, 2010
  • ACM SIGPLAN Notices
  • Mario Méndez-Lojo +7
  • Research Article
  • Citations1

Cache blocking for flux reconstruction: Extension to Navier-Stokes equations and anti-aliasing

  • Aug 05, 2024
  • Computer Physics Communications
  • Semih Akkurt +2
  • Book Chapter
  • Citations3

Hydrodynamical simulations of the universe

  • Jan 01, 1997
  • Parallel Computational Fluid Dynamics '96
  • F.R Pearce +8
  • Conference Article
  • Citations28

Tulip: a portable run-time system for object-parallel systems

  • Dec 23, 2002
  • P Beckman +1
  • Conference Article
  • Citations26

Automatically enhancing locality for tree traversals with traversal splicing

  • Oct 19, 2012
  • Youngjoon Jo +1
  • Research Article
  • Citations2

Automatically enhancing locality for tree traversals with traversal splicing

  • Oct 19, 2012
  • ACM SIGPLAN Notices
  • Youngjoon Jo +1
  • Single Book
  • Citations87

High-Performance Computing

  • Jan 01, 1999
  • R J Allan +4
  • Research Article
  • Citations10

Parallel explicit unstructured grid solvers on distributed memory computers

  • Aug 01, 1996
  • Advances in Engineering Software
  • Jean Cabello
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.