Standard
OpenCL task partitioning in the presence of GPU contention. / Grewe, Dominik
; Wang, Zheng; O'Boyle, Michael.
Languages and compilers for parallel computing: 26th International Workshop, LCPC 2013, San Jose, CA, USA, September 25--27, 2013. Revised Selected Papers. ed. / Călin Cașcaval; Pablo Montesinos . Springer, 2014. p. 87-101 (Lecture Notes in Computer Science; Vol. 8664).
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Harvard
Grewe, D
, Wang, Z & O'Boyle, M 2014,
OpenCL task partitioning in the presence of GPU contention. in C Cașcaval & P Montesinos (eds),
Languages and compilers for parallel computing: 26th International Workshop, LCPC 2013, San Jose, CA, USA, September 25--27, 2013. Revised Selected Papers. Lecture Notes in Computer Science, vol. 8664, Springer, pp. 87-101.
https://doi.org/10.1007/978-3-319-09967-5_5APA
Grewe, D.
, Wang, Z., & O'Boyle, M. (2014).
OpenCL task partitioning in the presence of GPU contention. In C. Cașcaval, & P. Montesinos (Eds.),
Languages and compilers for parallel computing: 26th International Workshop, LCPC 2013, San Jose, CA, USA, September 25--27, 2013. Revised Selected Papers (pp. 87-101). (Lecture Notes in Computer Science; Vol. 8664). Springer.
https://doi.org/10.1007/978-3-319-09967-5_5Vancouver
Grewe D
, Wang Z, O'Boyle M.
OpenCL task partitioning in the presence of GPU contention. In Cașcaval C, Montesinos P, editors, Languages and compilers for parallel computing: 26th International Workshop, LCPC 2013, San Jose, CA, USA, September 25--27, 2013. Revised Selected Papers. Springer. 2014. p. 87-101. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-09967-5_5
Author
Grewe, Dominik
; Wang, Zheng ; O'Boyle, Michael. /
OpenCL task partitioning in the presence of GPU contention. Languages and compilers for parallel computing: 26th International Workshop, LCPC 2013, San Jose, CA, USA, September 25--27, 2013. Revised Selected Papers. editor / Călin Cașcaval ; Pablo Montesinos . Springer, 2014. pp. 87-101 (Lecture Notes in Computer Science).
Bibtex
@inproceedings{59ae0aee5d2347c8b66d40c1d206c339,
title = "OpenCL task partitioning in the presence of GPU contention",
abstract = "Heterogeneous multi- and many-core systems are increasingly prevalent in the desktop and mobile domains. On these systems it is common for programs to compete with co-running programs for resources. While multi-task scheduling for CPUs is a well-studied area, how to partitioning and map computing tasks onto the heterogeneous system in the presence of GPU contention (i.e. multiple programs compete for the GPU) remains an outstanding problem.In this paper we consider the problem of partitioning OpenCL kernels on a CPU-GPU based system in the presence of contention on the GPU. We propose a machine learning-based approach that predicts the optimal partitioning of OpenCL kernels, explicitly taking GPU contention into account. Our predictive model achieves a speed-up of 1.92 over a scheme that always uses the GPU. When compared to two state-of-the-art dynamic approaches our model achieves speed-ups of 1.54 and 2.56 respectively.",
author = "Dominik Grewe and Zheng Wang and Michael O'Boyle",
year = "2014",
doi = "10.1007/978-3-319-09967-5_5",
language = "English",
isbn = "9783319099668",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "87--101",
editor = "Cașcaval, {C{\u a}lin } and {Montesinos }, Pablo",
booktitle = "Languages and compilers for parallel computing",
}
RIS
TY - GEN
T1 - OpenCL task partitioning in the presence of GPU contention
AU - Grewe, Dominik
AU - Wang, Zheng
AU - O'Boyle, Michael
PY - 2014
Y1 - 2014
N2 - Heterogeneous multi- and many-core systems are increasingly prevalent in the desktop and mobile domains. On these systems it is common for programs to compete with co-running programs for resources. While multi-task scheduling for CPUs is a well-studied area, how to partitioning and map computing tasks onto the heterogeneous system in the presence of GPU contention (i.e. multiple programs compete for the GPU) remains an outstanding problem.In this paper we consider the problem of partitioning OpenCL kernels on a CPU-GPU based system in the presence of contention on the GPU. We propose a machine learning-based approach that predicts the optimal partitioning of OpenCL kernels, explicitly taking GPU contention into account. Our predictive model achieves a speed-up of 1.92 over a scheme that always uses the GPU. When compared to two state-of-the-art dynamic approaches our model achieves speed-ups of 1.54 and 2.56 respectively.
AB - Heterogeneous multi- and many-core systems are increasingly prevalent in the desktop and mobile domains. On these systems it is common for programs to compete with co-running programs for resources. While multi-task scheduling for CPUs is a well-studied area, how to partitioning and map computing tasks onto the heterogeneous system in the presence of GPU contention (i.e. multiple programs compete for the GPU) remains an outstanding problem.In this paper we consider the problem of partitioning OpenCL kernels on a CPU-GPU based system in the presence of contention on the GPU. We propose a machine learning-based approach that predicts the optimal partitioning of OpenCL kernels, explicitly taking GPU contention into account. Our predictive model achieves a speed-up of 1.92 over a scheme that always uses the GPU. When compared to two state-of-the-art dynamic approaches our model achieves speed-ups of 1.54 and 2.56 respectively.
U2 - 10.1007/978-3-319-09967-5_5
DO - 10.1007/978-3-319-09967-5_5
M3 - Conference contribution/Paper
SN - 9783319099668
T3 - Lecture Notes in Computer Science
SP - 87
EP - 101
BT - Languages and compilers for parallel computing
A2 - Cașcaval, Călin
A2 - Montesinos , Pablo
PB - Springer
ER -
