Constant time worker thread allocation via configuration caching
Abstract
Mechanisms are provided for allocating threads for execution of a parallel region of code. A request for allocation of worker threads to execute the parallel region of code is received from a master thread. Cached thread allocation information identifying prior thread allocations that have been performed for the master thread are accessed. Worker threads are allocated to the master thread based on the cached thread allocation information. The parallel region of code is executed using the allocated worker threads.
- Inventors:
- Issue Date:
- Research Org.:
- International Business Machines Corp., Armonk, NY (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1163186
- Patent Number(s):
- 8881159
- Application Number:
- 13/070,811
- Assignee:
- International Business Machine Corporation (Armonk, NY)
- Patent Classifications (CPCs):
-
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
- DOE Contract Number:
- B554331
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 2011 Mar 24
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING
Citation Formats
Eichenberger, Alexandre E, and O'Brien, John K. P. Constant time worker thread allocation via configuration caching. United States: N. p., 2014.
Web.
Eichenberger, Alexandre E, & O'Brien, John K. P. Constant time worker thread allocation via configuration caching. United States.
Eichenberger, Alexandre E, and O'Brien, John K. P. Tue .
"Constant time worker thread allocation via configuration caching". United States. https://www.osti.gov/servlets/purl/1163186.
@article{osti_1163186,
title = {Constant time worker thread allocation via configuration caching},
author = {Eichenberger, Alexandre E and O'Brien, John K. P.},
abstractNote = {Mechanisms are provided for allocating threads for execution of a parallel region of code. A request for allocation of worker threads to execute the parallel region of code is received from a master thread. Cached thread allocation information identifying prior thread allocations that have been performed for the master thread are accessed. Worker threads are allocated to the master thread based on the cached thread allocation information. The parallel region of code is executed using the allocated worker threads.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2014},
month = {11}
}
Works referenced in this record:
MPI versus MPI+OpenMP on the IBM SP for the NAS Benchmarks
conference, January 2000
- Cappello, F.; Etiemble, D.
- ACM/IEEE SC 2000 Conference (SC'00)