Vector send operation for message-based communication
Abstract
Methods and systems for conducting vector send operations are provided. The processor of a sender node receives a request to perform a collective send operation (e.g., MPI_Broadcast) from a user application, requesting a copy of data in one or more send buffers by sent to each of a plurality of destinations in a destination vector. The processor invokes a vector send operation from a software communications library, placing a remote enqueue atomic send command for each destination node of the destination vector in an entry of a transmit data mover (XDM) command queue in a single call. The processor executes all of the commands in the XDM command queue and writes the data in the one or more send buffers into each receive queue of each destination identified in the destination vector.
- Inventors:
- Issue Date:
- Research Org.:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States); Hewlett Packard Enterprise Development LP, Houston, TX (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1860215
- Patent Number(s):
- 11210089
- Application Number:
- 16/508,769
- Assignee:
- Hewlett Packard Enterprise Development LP (Houston, TX)
- Patent Classifications (CPCs):
-
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
- DOE Contract Number:
- B621301
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 07/11/2019
- Country of Publication:
- United States
- Language:
- English
Citation Formats
Byrne, John L., Kuno, Harumi, and Drummond, Jeffrey. Vector send operation for message-based communication. United States: N. p., 2021.
Web.
Byrne, John L., Kuno, Harumi, & Drummond, Jeffrey. Vector send operation for message-based communication. United States.
Byrne, John L., Kuno, Harumi, and Drummond, Jeffrey. Tue .
"Vector send operation for message-based communication". United States. https://www.osti.gov/servlets/purl/1860215.
@article{osti_1860215,
title = {Vector send operation for message-based communication},
author = {Byrne, John L. and Kuno, Harumi and Drummond, Jeffrey},
abstractNote = {Methods and systems for conducting vector send operations are provided. The processor of a sender node receives a request to perform a collective send operation (e.g., MPI_Broadcast) from a user application, requesting a copy of data in one or more send buffers by sent to each of a plurality of destinations in a destination vector. The processor invokes a vector send operation from a software communications library, placing a remote enqueue atomic send command for each destination node of the destination vector in an entry of a transmit data mover (XDM) command queue in a single call. The processor executes all of the commands in the XDM command queue and writes the data in the one or more send buffers into each receive queue of each destination identified in the destination vector.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2021},
month = {12}
}
Works referenced in this record:
System and Method for Dynamic and Adaptive Interrupt Coalescing
patent-application, September 2018
- Benisty, Shay; Erez, Eran
- US Patent Application 15/468620; 20180275872
In-node aggregation and disaggregation of MPI alltoall and alltoallv collectives
patent-application, September 2017
- Shuler, Shahaf; Bloch, Noam; Bloch, Gil
- US Patent Application 15/446004; 20170255501
Remote direct memory access over datagrams
patent, December 2014
- Grant, Ryan; Rashti, Mohammad Javad; Balaji, Pavan
- US Patent Document 8,903,935
Shared Address Collectives Using Counter Mechanisms
patent-application, March 2011
- Blocksome, Michael; Dozsa, Gabor; Gooding, Thomas M.
- US Patent Application 12/568115; 20110078249
Methods and apparatus to improve performance data collection of a high performance computing application
patent-application, June 2019
- Ozog, David; Rahman, Md. Wasi-ur; Dinan, James
- US Patent Application 16/286095; 20190188111
Spatial and temporal merging of remote atomic operations
patent, February 2020
- Hughes, Christopher J.; Nuzman, Joseph; Svennebring, Jonas
- US Patent Document 10,572,260
Optimization of Collective Communication Operations in MPICH
journal, February 2005
- Thakur, Rajeev; Rabenseifner, Rolf; Gropp, William
- The International Journal of High Performance Computing Applications, Vol. 19, Issue 1
Intelligent data coordination for accelerated computing in cloud environment
patent, January 2021
- Zhao, Junping; Sun, Yifan; Peng, Layne Lin
- US Patent Document 10,891,156
Facilitating intra-node data transfer in collective communications
patent-application, April 2006
- Blackmore, Robert S.; Jia, Bin; Treumann, Richard R.
- US Patent Application 10/962721; 20060080668
Low Latency Multicast for Infiniband Host Channel Adapters
patent-application, March 2009
- Craddock, David; Gregg, Thomas A.
- US Patent Application 11/855562; 20090077268
Configuring compute nodes in a parallel computer using remote direct memory access ('RDMA')
patent-application, July 2013
- Aho, Michael E.; Attinella, John E.; Gooding, Thomas M.
- US Patent Application 13/709567; 20130185375
Device, system, and method of distributing messages
patent, August 2012
- Oved, Tzah
- US Patent Document 8,244,902
Dual-path computer interconnect system with four-ported packet memory control
patent, March 1993
- Awiszio, Desiree A.; Soman, Satish; Clark, Paul H.
- US Patent Document 5,193,149
Multi-Core Computer Processor
patent-application, September 2014
- Shalf, John; Donofrio, David; Oliker, Leonid
- US Patent Application 14/354257; 20140281243
High Performance RDMA-Based MPI Implementation over InfiniBand
journal, June 2004
- Liu, Jiuxing; Wu, Jiesheng; Panda, Dhabaleswar K.
- International Journal of Parallel Programming, Vol. 32, Issue 3