Collective network for computer structures
Abstract
A system and method for enabling high-speed, low-latency global collective communications among interconnected processing nodes. The global collective network optimally enables collective reduction operations to be performed during parallel algorithm operations executing in a computer structure having a plurality of the interconnected processing nodes. Router devices are included that interconnect the nodes of the network via links to facilitate performance of low-latency global processing operations at nodes of the virtual network. The global collective network may be configured to provide global barrier and interrupt functionality in asynchronous or synchronized manner. When implemented in a massively-parallel supercomputing structure, the global collective network is physically and logically partitionable according to the needs of a processing algorithm.
- Inventors:
- Issue Date:
- Research Org.:
- International Business Machines Corp., Armonk, NY (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1113767
- Patent Number(s):
- 8626957
- Application Number:
- 13/101,566
- Assignee:
- International Business Machines Corporation (Armonk, NY)
- Patent Classifications (CPCs):
-
H - ELECTRICITY H03 - BASIC ELECTRONIC CIRCUITRY H03M - CODING
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
- DOE Contract Number:
- B517552
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 2011 May 05
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING
Citation Formats
Blumrich, Matthias A, Coteus, Paul W, Chen, Dong, Gara, Alan, Giampapa, Mark E, Heidelberger, Philip, Hoenicke, Dirk, Takken, Todd E, Steinmacher-Burow, Burkhard D, and Vranas, Pavlos M. Collective network for computer structures. United States: N. p., 2014.
Web.
Blumrich, Matthias A, Coteus, Paul W, Chen, Dong, Gara, Alan, Giampapa, Mark E, Heidelberger, Philip, Hoenicke, Dirk, Takken, Todd E, Steinmacher-Burow, Burkhard D, & Vranas, Pavlos M. Collective network for computer structures. United States.
Blumrich, Matthias A, Coteus, Paul W, Chen, Dong, Gara, Alan, Giampapa, Mark E, Heidelberger, Philip, Hoenicke, Dirk, Takken, Todd E, Steinmacher-Burow, Burkhard D, and Vranas, Pavlos M. Tue .
"Collective network for computer structures". United States. https://www.osti.gov/servlets/purl/1113767.
@article{osti_1113767,
title = {Collective network for computer structures},
author = {Blumrich, Matthias A and Coteus, Paul W and Chen, Dong and Gara, Alan and Giampapa, Mark E and Heidelberger, Philip and Hoenicke, Dirk and Takken, Todd E and Steinmacher-Burow, Burkhard D and Vranas, Pavlos M},
abstractNote = {A system and method for enabling high-speed, low-latency global collective communications among interconnected processing nodes. The global collective network optimally enables collective reduction operations to be performed during parallel algorithm operations executing in a computer structure having a plurality of the interconnected processing nodes. Router devices are included that interconnect the nodes of the network via links to facilitate performance of low-latency global processing operations at nodes of the virtual network. The global collective network may be configured to provide global barrier and interrupt functionality in asynchronous or synchronized manner. When implemented in a massively-parallel supercomputing structure, the global collective network is physically and logically partitionable according to the needs of a processing algorithm.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2014},
month = {1}
}