Skip to main content
USENIX
  • Conferences
  • Students
Sign in
  • HotPar '12 Home
  • Registration and Lodging
  • Organizers
  • Workshop Program
  • Poster Session
  • Birds-of-a-Feather Sessions
  • Travel
  • Calendar
  • Students
  • Questions?
  • For Participants
  • Call for Papers
  • Past Proceedings

sponsors

Gold Sponsor
Bronze Sponsor
Bronze Sponsor
Bronze Sponsor

twitter

Tweets by @usenix

usenix conference policies

  • Event Code of Conduct
  • Conference Network Policy
  • Statement on Environmental Responsibility Policy

You are here

Home » Performance Gains in Conjugate Gradient Computation with Linearly Connected GPU Multiprocessors
Tweet

connect with us

http://twitter.com/usenix
http://www.facebook.com/usenixassociation

Performance Gains in Conjugate Gradient Computation with Linearly Connected GPU Multiprocessors

Authors: 

Stephen J. Tarsa, Tsung-han Lin, and H.T. Kung, Harvard University

Abstract: 

Conjugate gradient is an important iterative method used for solving least squares problems. It is compute-bound and generally involves only simple matrix computations. One would expect that we could fully parallelize such computation on the GPU architecture with multiple Stream Multiprocessors (SMs), each consisting of many SIMD processing units. While implementing a conjugate gradient method for compressive sensing signal reconstruction, we have noticed that large speed-up due to parallel processing is actually infeasible due to the high I/O cost between SMs and GPU global memory. We have found that if SMs were linearly connected, we could gain a 15x speedup by loop unrolling. We conclude that adding these relatively inexpensive neighbor connections for SMs can significantly enhance the applicability of GPUs to a large class of similar matrix computations.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@conference {259241,
title = {Performance Gains in Conjugate Gradient Computation with Linearly Connected {GPU} Multiprocessors},
year = {2012},
address = {Berkeley, CA},
publisher = {USENIX Association},
month = jun,
}
Download
Tarsa PDF
  • Log in or    Register to post comments

Gold Sponsors

Bronze Sponsors

© USENIX

  • Privacy Policy
  • Contact Us