Exploring Machine Learning for Thread Characterization on Heterogeneous Multiprocessors
ACM SIGOPS Operating Systems Review · 11.09.2017 · 51 (1) · 113-123
DOI: 10.1145/3139645.3139664 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM SIGOPS Operating Systems Review, online erschienen am 11.09.2017. Verfasst von Cha V. Li und 2 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 31 Literaturangaben.
Abstract (Originalsprache)
We introduce a thread characterization method that explores hardware performance counters and machine learning techniques to automate estimating workload execution on heterogeneous processors. We show that our characterization scheme achieves higher accuracy when predicting performance indicators, such as instructions per cycle and last-level cache misses, commonly used to determine the mapping of threads to processor types at runtime. We also show that support vector regression achieves higher accuracy when compared to linear regression, and has very low (1%) overhead. The results presented in this paper can provide a foundation for advanced investigations and interesting new directions in intelligent thread scheduling and power management on multiprocessors.
Zitieren
APA
Li, C. V., Petrucci, V., & Mossé, D. (2017). Exploring Machine Learning for Thread Characterization on Heterogeneous Multiprocessors. ACM SIGOPS Operating Systems Review, 51(1), 113-123. https://doi.org/10.1145/3139645.3139664
Vancouver
Li CV, Petrucci V, Mossé D. Exploring Machine Learning for Thread Characterization on Heterogeneous Multiprocessors. ACM SIGOPS Operating Systems Review. 2017;51(1):113-123. doi:10.1145/3139645.3139664
BibTeX
@article{Li2017,
author = {Li, Cha V. and Petrucci, Vinicius and Mossé, Daniel},
title = {Exploring Machine Learning for Thread Characterization on Heterogeneous Multiprocessors},
journal = {ACM SIGOPS Operating Systems Review},
year = {2017},
volume = {51},
number = {1},
pages = {113--123},
doi = {10.1145/3139645.3139664}
}
EUVIMED