loading...
FastForward for Efficient Pipeline Parallelism
Brasov, Romania September 15-September 19
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/PACT.2007.3616th International Conference on Para ...
 This Article 
 
PURCHASE ARTICLE: $0
HTML
IEEE Xplore Subscribers
 
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
ASCII Text
x 
 
John Giacomoni, Tipp Moseley, Manish Vachharajani, "FastForward for Efficient Pipeline Parallelism," Parallel Architectures and Compilation Techniques, International Conference on, pp. 407, 16th International Conference on Parallel Architecture and Compilation Techniques (PACT 2007), 2007.
 
 
BibTex
x
 
@article{ 10.1109/PACT.2007.36,
author = {John Giacomoni and Tipp Moseley and Manish Vachharajani},
title = {FastForward for Efficient Pipeline Parallelism},
journal ={Parallel Architectures and Compilation Techniques, International Conference on},
volume = {0},
year = {2007},
issn = {1089-795X},
pages = {407},
doi = {http://doi.ieeecomputersociety.org/10.1109/PACT.2007.36},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
}
 
 
RefWorks Procite/RefMan/Endnote
x 
 
TY - CONF
JO - Parallel Architectures and Compilation Techniques, International Conference on
TI - FastForward for Efficient Pipeline Parallelism
SN - 1089-795X
SP
EP
A1 - John Giacomoni,
A1 - Tipp Moseley,
A1 - Manish Vachharajani,
PY - 2007
KW - null
VL - 0
JA - Parallel Architectures and Compilation Techniques, International Conference on
ER -
 
John Giacomoni, University of Colorado at Boulder, USA
Tipp Moseley, University of Colorado at Boulder, USA
Manish Vachharajani, University of Colorado at Boulder, USA
High-rate core-to-core communication is critical for efficient pipeline-parallel software architectures. This paper introduces FastForward, a software-only low-overhead high-rate queue algorithm for pipeline parallelism on multicore architectures. FastForward uses an architecturallytuned domain-specific adaptation of concurrent lock-free queues to provide low-latency and low-overhead core-to-core communication. Enqueue and dequeue times on a 2 GHz Opteron 270 based system are as low as 36 ns, up to 4x faster than Lamport?s solution.
Citation:
John Giacomoni, Tipp Moseley, Manish Vachharajani, "FastForward for Efficient Pipeline Parallelism," pact, pp.407, 16th International Conference on Parallel Architecture and Compilation Techniques (PACT 2007), 2007
Usage of this product signifies your acceptance of the Terms of Use.