Collective communications apparatus and method for parallel systems
Abstract
A collective communication apparatus and method for parallel computing systems. For example, one embodiment of an apparatus comprises a plurality of processor elements (PEs); collective interconnect logic to dynamically form a virtual collective interconnect (VCI) between the PEs at runtime without global communication among all of the PEs, the VCI defining a logical topology between the PEs in which each PE is directly communicatively coupled to a only a subset of the remaining PEs; and execution logic to execute collective operations across the PEs, wherein one or more of the PEs receive first results from a first portion of the subset of the remaining PEs, perform a portion of the collective operations, and provide second results to a second portion of the subset of the remaining PEs.
- Inventors:
- Issue Date:
- Research Org.:
- Intel Corporation, Santa Clara, CA (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1330323
- Patent Number(s):
- 9477628
- Application Number:
- 14/040,676
- Assignee:
- Intel Corporation (Santa Clara, CA)
- Patent Classifications (CPCs):
-
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
- DOE Contract Number:
- B600738
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 2013 Sep 28
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING
Citation Formats
Knies, Allan D., Keppel, David Pardo, Woo, Dong Hyuk, and Fryman, Joshua B. Collective communications apparatus and method for parallel systems. United States: N. p., 2016.
Web.
Knies, Allan D., Keppel, David Pardo, Woo, Dong Hyuk, & Fryman, Joshua B. Collective communications apparatus and method for parallel systems. United States.
Knies, Allan D., Keppel, David Pardo, Woo, Dong Hyuk, and Fryman, Joshua B. Tue .
"Collective communications apparatus and method for parallel systems". United States. https://www.osti.gov/servlets/purl/1330323.
@article{osti_1330323,
title = {Collective communications apparatus and method for parallel systems},
author = {Knies, Allan D. and Keppel, David Pardo and Woo, Dong Hyuk and Fryman, Joshua B.},
abstractNote = {A collective communication apparatus and method for parallel computing systems. For example, one embodiment of an apparatus comprises a plurality of processor elements (PEs); collective interconnect logic to dynamically form a virtual collective interconnect (VCI) between the PEs at runtime without global communication among all of the PEs, the VCI defining a logical topology between the PEs in which each PE is directly communicatively coupled to a only a subset of the remaining PEs; and execution logic to execute collective operations across the PEs, wherein one or more of the PEs receive first results from a first portion of the subset of the remaining PEs, perform a portion of the collective operations, and provide second results to a second portion of the subset of the remaining PEs.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2016},
month = {10}
}
Works referenced in this record:
Arithmetic unit for SIMD type parallel computer
patent, May 1995
- Matuda, Motohiko; Yuasa, Taiichi
- US Patent Document 5,418,915
Parallel computer with reconstruction of processor clusters
patent, July 1997
- Tsujimichi, Shingo; Kondo, Michimasa
- US Patent Document 5,649,106
Distributed shared memory multiprocessor and data processing method
patent, June 2013
- Akaike, Yukihiko; Suzuki, Hitoshi
- US Patent Document 8,458,411
Executing a Gather Operation on a Parallel Computer
patent-application, October 2010
- Archer, Charles J.; Ratterman, Joseph D.
- US Patent Application 11/754740; 20100274997