skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Performing optimized collective operations in an irregular subcommunicator of compute nodes in a parallel computer

Abstract

In a parallel computer, performing optimized collective operations in an irregular subcommunicator of compute nodes may be carried out by: identifying, within the irregular subcommunicator, regular neighborhoods of compute nodes; selecting, for each neighborhood from the compute nodes of the neighborhood, a local root node; assigning each local root node to a node of a neighborhood-wide tree topology; mapping, for each neighborhood, the compute nodes of the neighborhood to a local tree topology having, at its root, the local root node of the neighborhood; and performing a one way, rooted collective operation within the subcommunicator including: performing, in one phase, the collective operation within each neighborhood; and performing, in another phase, the collective operation amongst the local root nodes.

Inventors:
;
Publication Date:
Research Org.:
International Business Machines Corp., Armonk, NY (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1568731
Patent Number(s):
10,382,527
Application Number:
14/055,402
Assignee:
International Business Machines Corporation (Armonk, NY)
DOE Contract Number:  
B554431
Resource Type:
Patent
Resource Relation:
Patent File Date: 10/16/2013
Country of Publication:
United States
Language:
English

Citation Formats

Davis, Kristan D., and Faraj, Daniel A. Performing optimized collective operations in an irregular subcommunicator of compute nodes in a parallel computer. United States: N. p., 2019. Web.
Davis, Kristan D., & Faraj, Daniel A. Performing optimized collective operations in an irregular subcommunicator of compute nodes in a parallel computer. United States.
Davis, Kristan D., and Faraj, Daniel A. Tue . "Performing optimized collective operations in an irregular subcommunicator of compute nodes in a parallel computer". United States. https://www.osti.gov/servlets/purl/1568731.
@article{osti_1568731,
title = {Performing optimized collective operations in an irregular subcommunicator of compute nodes in a parallel computer},
author = {Davis, Kristan D. and Faraj, Daniel A.},
abstractNote = {In a parallel computer, performing optimized collective operations in an irregular subcommunicator of compute nodes may be carried out by: identifying, within the irregular subcommunicator, regular neighborhoods of compute nodes; selecting, for each neighborhood from the compute nodes of the neighborhood, a local root node; assigning each local root node to a node of a neighborhood-wide tree topology; mapping, for each neighborhood, the compute nodes of the neighborhood to a local tree topology having, at its root, the local root node of the neighborhood; and performing a one way, rooted collective operation within the subcommunicator including: performing, in one phase, the collective operation within each neighborhood; and performing, in another phase, the collective operation amongst the local root nodes.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2019},
month = {8}
}

Patent:

Save / Share:

Works referenced in this record:

Method for deadlock-free message passing in MIMD systems using routers and buffers
patent, January 1999


Virtual channel assignment in large torus systems
patent, August 2000


Mapping of nodes in an interconnection fabric
patent, February 2006


Method and apparatus for self-organizing node groups on a network
patent, December 2008


High performance storage array interconnection fabric using multiple independent paths
patent, July 2008


Storage array interconnection fabric using a torus topology
patent, April 2004