The University of Southampton
University of Southampton Institutional Repository

Selective policies for efficient state retention in transiently-powered embedded systems: exploiting properties of NVM technologies

Selective policies for efficient state retention in transiently-powered embedded systems: exploiting properties of NVM technologies
Selective policies for efficient state retention in transiently-powered embedded systems: exploiting properties of NVM technologies
Transiently-powered embedded systems are emerging to enable computation to be sustained during intermittent supply, without the need for large energy buffers such as batteries or supercapacitors. To deal with the intermittent nature of the input source, these systems save the system state (i.e. registers and main memory) to Non-Volatile Memory (NVM) before a power failure, and restore it when the power supply recovers. Existing approaches normally save the entire state of the system upon power failure, but this is both energy and time consuming. In this paper, we analyse existing approaches to identify their inefficiency when used with specific NVM technologies, and propose novel selective policies for efficiently retaining the system state by exploiting properties of different NVM technologies. These policies are based on (1) concatenating multiple images into the available NVM before erasing, and (2) efficiently selecting only the system state that has changed since last saving. The existing and proposed policies are experimentally validated on two embedded platforms featuring different NVM technologies (Flash and FRAM), depending on their characteristics, in order to identify the most energy efficient policy/platform combination. Results show a reduction in energy and time overhead of up to 90.6% for Flash memory using a novel policy, and 86.2% for FRAM, compared to the typical approach of saving the entire system state.
state retention, energy harvesting, Non-volatile memories, Transient computing
2210-5379
Verykios, Theodoros D.
fc203333-af9c-48e6-b7d6-f22d8cf60636
Balsamo, Domenico
fa2dc20a-e3da-4d74-9070-9c61c6a471ba
Merrett, Geoff V.
89b3a696-41de-44c3-89aa-b0aa29f54020
Verykios, Theodoros D.
fc203333-af9c-48e6-b7d6-f22d8cf60636
Balsamo, Domenico
fa2dc20a-e3da-4d74-9070-9c61c6a471ba
Merrett, Geoff V.
89b3a696-41de-44c3-89aa-b0aa29f54020

Verykios, Theodoros D., Balsamo, Domenico and Merrett, Geoff V. (2018) Selective policies for efficient state retention in transiently-powered embedded systems: exploiting properties of NVM technologies. Sustainable Computing: Informatics and Systems. (doi:10.1016/j.suscom.2018.07.003).

Record type: Article

Abstract

Transiently-powered embedded systems are emerging to enable computation to be sustained during intermittent supply, without the need for large energy buffers such as batteries or supercapacitors. To deal with the intermittent nature of the input source, these systems save the system state (i.e. registers and main memory) to Non-Volatile Memory (NVM) before a power failure, and restore it when the power supply recovers. Existing approaches normally save the entire state of the system upon power failure, but this is both energy and time consuming. In this paper, we analyse existing approaches to identify their inefficiency when used with specific NVM technologies, and propose novel selective policies for efficiently retaining the system state by exploiting properties of different NVM technologies. These policies are based on (1) concatenating multiple images into the available NVM before erasing, and (2) efficiently selecting only the system state that has changed since last saving. The existing and proposed policies are experimentally validated on two embedded platforms featuring different NVM technologies (Flash and FRAM), depending on their characteristics, in order to identify the most energy efficient policy/platform combination. Results show a reduction in energy and time overhead of up to 90.6% for Flash memory using a novel policy, and 86.2% for FRAM, compared to the typical approach of saving the entire system state.

Text Sustainable_tdv - Accepted Manuscript
Restricted to Repository staff only until 11 October 2019.
Request a copy
Text 1-s2.0-S2210537917304365-main - Proof
Available under License Creative Commons Attribution.
Download (3MB)

More information

Accepted/In Press date: 11 July 2018
e-pub ahead of print date: 25 July 2018
Keywords: state retention, energy harvesting, Non-volatile memories, Transient computing

Identifiers

Local EPrints ID: 422555
URI: https://eprints.soton.ac.uk/id/eprint/422555
ISSN: 2210-5379
PURE UUID: a3367573-6006-4818-bf7c-951f3e0a77c4
ORCID for Geoff V. Merrett: ORCID iD orcid.org/0000-0003-4980-3894

Catalogue record

Date deposited: 25 Jul 2018 16:30
Last modified: 27 Oct 2018 00:33

Export record

Altmetrics

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×