loading...
 This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 9
An Evaluation of Two Implementation Strategies for Optimizing One-Sided Atomic Reduction
Denver, Colorado
April 04-April 08
ISBN: 0-7695-2312-9
ASCII Text
x 
 
Jarek Nieplocha, Vinod Tipparaju, Edoardo Apra, "An Evaluation of Two Implementation Strategies for Optimizing One-Sided Atomic Reduction," Parallel and Distributed Processing Symposium, International, vol. 10, pp. 215b, 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 9, 2005.
 
 
BibTex
x
 
@article{ 10.1109/IPDPS.2005.96,
author = {Jarek Nieplocha and Vinod Tipparaju and Edoardo Apra},
title = {An Evaluation of Two Implementation Strategies for Optimizing One-Sided Atomic Reduction},
journal ={Parallel and Distributed Processing Symposium, International},
volume = {10},
year = {2005},
issn = {1530-2075},
pages = {215b},
doi = {http://doi.ieeecomputersociety.org/10.1109/IPDPS.2005.96},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
}
 
 
RefWorks Procite/RefMan/Endnote
x 
 
TY - CONF
JO - Parallel and Distributed Processing Symposium, International
TI - An Evaluation of Two Implementation Strategies for Optimizing One-Sided Atomic Reduction
SN - 1530-2075
SP
EP
A1 - Jarek Nieplocha,
A1 - Vinod Tipparaju,
A1 - Edoardo Apra,
PY - 2005
KW - null
VL - 10
JA - Parallel and Distributed Processing Symposium, International
ER -
 
Jarek Nieplocha, Pacific Northwest National Laboratory
Vinod Tipparaju, Pacific Northwest National Laboratory
Edoardo Apra, Pacific Northwest National Laboratory
One-sided atomic reduction, also known as the accumulate operation, combines atomically a content of the local buffer with data at remote memory location. This operation has been included in the MPI-2 standard a, MPI_Accumulate. The current paper discusses two strategies for implementing one-sided atomic reduction called owner-computes and callercomputes. Performance of these two schemes has been investigated on the HP Alphaserver SC45 and HP zx-2600 clusters both equipped with the Quadrics Elan-3 network.
Citation:
Jarek Nieplocha, Vinod Tipparaju, Edoardo Apra, "An Evaluation of Two Implementation Strategies for Optimizing One-Sided Atomic Reduction," ipdps, vol. 10, pp.215b, 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 9, 2005
Usage of this product signifies your acceptance of the Terms of Use.