Home // CENICS 2017, The Tenth International Conference on Advances in Circuits, Electronics and Micro-electronics // View article
Authors:
Thomas Jerabek
Benjamin Aigner
Florian Gerstmayer
Jürgen Hausladen
Keywords: dynamic frequency scaling; worst-case execution time analysis; energy-efficient computing
Abstract:
Embedded systems are at the core of many new emerging technologies and applications, deeply integrated into our daily lives.Especially, the demand for battery-powered solutions in consumer-related applications is growing, to support different environments and fields of application. Therefore, energy efficiency measures for embedded systems become even more important. In this paper, a dynamic frequency scaling approach for embedded systems is presented to reduce the overall energy consumption while still meeting time constraints within a real-time operating system. Starting with a general discussion and mathematical derivation along with an elaboration of the state of the art, our concept and implementation is discussed. This includes primarily the developed Worst-Case Execution Time (WCET) aware Earliest Deadline First (EDF) scheduler which is used to dynamically scale the frequency at runtime. Moreover, a use case targeting a real-time smart home application is provided, which was used to evaluate and compare our implementation in regard to it's energy consumption. The respective results are elaborated alongside possible future work and improvements.
Pages: 15 to 20
Copyright: Copyright (c) IARIA, 2017
Publication date: September 10, 2017
Published in: conference
ISSN: 2308-426X
ISBN: 978-1-61208-585-2
Location: Rome, Italy
Dates: from September 10, 2017 to September 14, 2017