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

Title: Optimizing software-directed instruction replication for GPU error detection

Abstract

Software-only and software-hardware optimizations to reduce the overhead of intra-thread instruction duplication on a GPU or other instruction processor are disclosed. The optimizations trade off error containment for performance and include ISA extensions with limited hardware changes and area costs.

Inventors:
; ; ; ;
Issue Date:
Research Org.:
NVIDIA Corp., Santa Clara, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1771561
Patent Number(s):
10817289
Application Number:
16/150,410
Assignee:
NVIDIA Corporation (Santa Clara, CA)
Patent Classifications (CPCs):
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
DOE Contract Number:  
B620719
Resource Type:
Patent
Resource Relation:
Patent File Date: 10/03/2018
Country of Publication:
United States
Language:
English

Citation Formats

Hari, Siva, Sullivan, Michael, Tsai, Timothy, Keckler, Stephen W., and Mahmoud, Abdulrahman. Optimizing software-directed instruction replication for GPU error detection. United States: N. p., 2020. Web.
Hari, Siva, Sullivan, Michael, Tsai, Timothy, Keckler, Stephen W., & Mahmoud, Abdulrahman. Optimizing software-directed instruction replication for GPU error detection. United States.
Hari, Siva, Sullivan, Michael, Tsai, Timothy, Keckler, Stephen W., and Mahmoud, Abdulrahman. Tue . "Optimizing software-directed instruction replication for GPU error detection". United States. https://www.osti.gov/servlets/purl/1771561.
@article{osti_1771561,
title = {Optimizing software-directed instruction replication for GPU error detection},
author = {Hari, Siva and Sullivan, Michael and Tsai, Timothy and Keckler, Stephen W. and Mahmoud, Abdulrahman},
abstractNote = {Software-only and software-hardware optimizations to reduce the overhead of intra-thread instruction duplication on a GPU or other instruction processor are disclosed. The optimizations trade off error containment for performance and include ISA extensions with limited hardware changes and area costs.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2020},
month = {10}
}