DOE Patents title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Network packet templating for GPU-initiated communication

Abstract

Systems, apparatuses, and methods for performing network packet templating for graphics processing unit (GPU)-initiated communication are disclosed. A central processing unit (CPU) creates a network packet according to a template and populates a first subset of fields of the network packet with static data. Next, the CPU stores the network packet in a memory. A GPU initiates execution of a kernel and detects a network communication request within the kernel and prior to the kernel completing execution. Responsive to this determination, the GPU populates a second subset of fields of the network packet with runtime data. Then, the GPU generates a notification that the network packet is ready to be processed. A network interface controller (NIC) processes the network packet using data retrieved from the first subset of fields and from the second subset of fields responsive to detecting the notification.

Inventors:
; ;
Issue Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1735137
Patent Number(s):
10740163
Application Number:
16/022,498
Assignee:
Advanced Micro Devices, Inc. (Santa Clara, CA)
Patent Classifications (CPCs):
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
H - ELECTRICITY H04 - ELECTRIC COMMUNICATION TECHNIQUE H04L - TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
DOE Contract Number:  
AC52-07NA27344
Resource Type:
Patent
Resource Relation:
Patent File Date: 06/28/2018
Country of Publication:
United States
Language:
English

Citation Formats

Hamidouche, Khaled, Lebeane, Michael Wayne, and Benton, Walter B. Network packet templating for GPU-initiated communication. United States: N. p., 2020. Web.
Hamidouche, Khaled, Lebeane, Michael Wayne, & Benton, Walter B. Network packet templating for GPU-initiated communication. United States.
Hamidouche, Khaled, Lebeane, Michael Wayne, and Benton, Walter B. Tue . "Network packet templating for GPU-initiated communication". United States. https://www.osti.gov/servlets/purl/1735137.
@article{osti_1735137,
title = {Network packet templating for GPU-initiated communication},
author = {Hamidouche, Khaled and Lebeane, Michael Wayne and Benton, Walter B.},
abstractNote = {Systems, apparatuses, and methods for performing network packet templating for graphics processing unit (GPU)-initiated communication are disclosed. A central processing unit (CPU) creates a network packet according to a template and populates a first subset of fields of the network packet with static data. Next, the CPU stores the network packet in a memory. A GPU initiates execution of a kernel and detects a network communication request within the kernel and prior to the kernel completing execution. Responsive to this determination, the GPU populates a second subset of fields of the network packet with runtime data. Then, the GPU generates a notification that the network packet is ready to be processed. A network interface controller (NIC) processes the network packet using data retrieved from the first subset of fields and from the second subset of fields responsive to detecting the notification.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2020},
month = {8}
}

Works referenced in this record:

dCUDA: Hardware Supported Overlap of Computation and Communication
conference, November 2016

  • Gysi, Tobias; Bar, Jeremia; Hoefler, Torsten
  • SC16: International Conference for High Performance Computing, Networking, Storage and Analysis
  • https://doi.org/10.1109/SC.2016.51

Infiniband-Verbs on GPU: A Case Study of Controlling an Infiniband Network Device from the GPU
conference, May 2014


Packet processing approach to improve performance and energy efficiency for software routers
patent, September 2016


Gravel: fine-grain GPU-initiated network messages
conference, November 2017

  • Orr, Marc S.; Che, Shuai; Beckmann, Bradford M.
  • SC '17: The International Conference for High Performance Computing, Networking, Storage and Analysis, Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
  • https://doi.org/10.1145/3126908.3126914