Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units
Abstract
A compute unit configured to execute multiple threads in parallel is presented. The compute unit includes one or more single instruction multiple data (SIMD) units and a fetch and decode logic. The SIMD units have differing numbers of arithmetic logic units (ALUs), such that each SIMD unit can execute a different number of threads. The fetch and decode logic is in communication with each of the SIMD units, and is configured to assign the threads to the SIMD units for execution based on such differing numbers of ALUs.
- Inventors:
- Issue Date:
- Research Org.:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1735025
- Patent Number(s):
- 10713059
- Application Number:
- 14/490,213
- Assignee:
- Advanced Micro Devices, Inc. (Sunnyvale, CA)
- Patent Classifications (CPCs):
-
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
- DOE Contract Number:
- AC52-07NA27344; B600716
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 09/18/2014
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING
Citation Formats
Greathouse, Joseph L., Meswani, Mitesh R., Puthoor, Sooraj, Yudanov, Dmitri, and O'Connor, James M. Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units. United States: N. p., 2020.
Web.
Greathouse, Joseph L., Meswani, Mitesh R., Puthoor, Sooraj, Yudanov, Dmitri, & O'Connor, James M. Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units. United States.
Greathouse, Joseph L., Meswani, Mitesh R., Puthoor, Sooraj, Yudanov, Dmitri, and O'Connor, James M. Tue .
"Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units". United States. https://www.osti.gov/servlets/purl/1735025.
@article{osti_1735025,
title = {Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units},
author = {Greathouse, Joseph L. and Meswani, Mitesh R. and Puthoor, Sooraj and Yudanov, Dmitri and O'Connor, James M.},
abstractNote = {A compute unit configured to execute multiple threads in parallel is presented. The compute unit includes one or more single instruction multiple data (SIMD) units and a fetch and decode logic. The SIMD units have differing numbers of arithmetic logic units (ALUs), such that each SIMD unit can execute a different number of threads. The fetch and decode logic is in communication with each of the SIMD units, and is configured to assign the threads to the SIMD units for execution based on such differing numbers of ALUs.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2020},
month = {7}
}
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.
Works referenced in this record:
Techniques for Increasing Vector Processing Utilization and Efficiency Through Vector Lane Predication Prediction
patent-application, May 2015
- Lee, Hung Q.; Moreira, Jose E.; Pattnaik, Pratap C.
- US Patent Application 14/082652; 20150143083
Digital signal processor and baseband communication device
patent-application, November 2014
- Nilsson, Anders; Tell, Eric
- US Patent Application 14/364629; 20140344549
Monitoring Vector Lane Duty Cycle for Dynamic Optimization
patent-application, August 2015
- Kim, Daehyun; Park, Jong Soo; Woo, Dong Hyuk
- US Patent Application 14/190404; 20150242210
Temporal SIMT Execution Optimization
patent-application, February 2013
- Krashinsky, Ronny M.
- US Patent Application 13/209189; 20130042090
Clock control of functional units in an integrated circuit based on monitoring unit signals to predict inactivity
patent, January 2006
- Filippo, Michael
- US Patent Document 6,983,389
Device and Method for Managing SIMD Architecture Based Thread Divergence
patent-application, May 2016
- Jin, Seung-Hun
- US Patent Application 14/786417; 20160132338
Compressing Execution Cycles for Divergent Execution in a Single Instruction Multiple Data (SIMD) Processor
patent-application, June 2014
- Vaidya, Aniruddha S.; Shayesteh, Anahita; Woo, Doug Hyuk
- US Patent Application 13/724633; 20140181477
Variable width vector instruction processor
patent-application, December 2011
- Karkhanis, Tejas; Moreira, Jose E.; Salapura, Valentina
- US Patent Application 12/825328; 20110320765
Structured programming control flow using a disable mask in a SIMD architecture
patent, November 2009
- Coon, Brett W.; Lindholm, John Erik; Tzvetkov, Svetoslav D.
- US Patent Document 7,617,384
Scalable computing array
patent-application, May 2014
- Krig, Scott
- US Patent Application 13/672828; 20140136816