Fine-Grained Power Modeling of Multicore Processors Using FFNNs

Mark Sagi, Nguyen Anh Vu Doan, Nael Fasfous, Thomas Wild, Andreas Herkersdorf

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

4 Zitate (Scopus)

Abstract

To minimize power consumption while maximizing performance, today’s multicore processors rely on fine-grained run-time dynamic power information – both in the time domain, e.g. μs to ms, and space domain, e.g. core-level. The state-of-the-art for deriving such power information is mainly based on predetermined power models which use linear modeling techniques to determine the core-performance/core-power relationship. However, with multicore processors becoming ever more complex, linear modeling techniques cannot capture all possible core-performance related power states anymore. Although, artificial neural networks (ANN) have been proposed for coarse-grained power modeling of servers with time resolutions in the range of seconds, no work has yet investigated fine-grained ANN-based power modeling. In this paper, we explore feed-forward neural networks (FFNNs) for core-level power modeling with estimation rates in the range of 10 kHz. To achieve a high estimation accuracy, we determine optimized neural network architectures and train FFNNs on performance counter and power data from a complex-out-of-order processor architecture. We show that, relative power estimation error decreases on average by 7.5% compared to a state-of-the-art linear power modeling approach and decreases by 5.5% compared to a multivariate polynomial regression model. Furthermore, we propose an implementation for run-time inference of the power modeling FFNN and show that the area overhead is negligible.

OriginalspracheEnglisch
TitelEmbedded Computer Systems
UntertitelArchitectures, Modeling, and Simulation - 20th International Conference, SAMOS 2020, Proceedings
Redakteure/-innenAlex Orailoglu, Matthias Jung, Marc Reichenbach
Herausgeber (Verlag)Springer Science and Business Media Deutschland GmbH
Seiten186-199
Seitenumfang14
ISBN (Print)9783030609382
DOIs
PublikationsstatusVeröffentlicht - 2020
Veranstaltung20th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation, SAMOS 2020 - Samos, Griechenland
Dauer: 5 Juli 20209 Juli 2020

Publikationsreihe

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Band12471 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Konferenz

Konferenz20th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation, SAMOS 2020
Land/GebietGriechenland
OrtSamos
Zeitraum5/07/209/07/20

Fingerprint

Untersuchen Sie die Forschungsthemen von „Fine-Grained Power Modeling of Multicore Processors Using FFNNs“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren