Efficient state retention through paged memory management for reactive transient computing
Efficient state retention through paged memory management for reactive transient computing
Reactive transient computing systems preserve computational progress despite frequent power failures by suspending (saving state to nonvolatile memory) when detecting a power failure, and restoring once power returns. Existing methods inefficiently save and restore all allocated memory. We propose lightweight memory management that applies the concept of paging to load pages only when needed, and save only modified pages. We then develop a model that maximises available execution time by dynamically adjusting the suspend and restore voltage thresholds. Experiments on an MSP430FR5994 microcontroller show that our method reduces state retention overheads by up to 86.9% and executes algorithms up to 5.3 times faster than the state-of-the-art.
Transient Computing, Internet of Things, Batteryless Computing
Sliper, Sivert T.
73303db3-fb3d-4434-973b-3def05837e7f
Balsamo, Domenico
fa2dc20a-e3da-4d74-9070-9c61c6a471ba
Nikoleris, Nikos
be54f3c1-c36e-4dde-8611-0af54b56e033
Wang, William
ff81a455-8a66-49db-82f2-849bc6dc2c51
Weddell, Alexander
3d8c4d63-19b1-4072-a779-84d487fd6f03
Merrett, Geoff
89b3a696-41de-44c3-89aa-b0aa29f54020
6 June 2019
Sliper, Sivert T.
73303db3-fb3d-4434-973b-3def05837e7f
Balsamo, Domenico
fa2dc20a-e3da-4d74-9070-9c61c6a471ba
Nikoleris, Nikos
be54f3c1-c36e-4dde-8611-0af54b56e033
Wang, William
ff81a455-8a66-49db-82f2-849bc6dc2c51
Weddell, Alexander
3d8c4d63-19b1-4072-a779-84d487fd6f03
Merrett, Geoff
89b3a696-41de-44c3-89aa-b0aa29f54020
Sliper, Sivert T., Balsamo, Domenico, Nikoleris, Nikos, Wang, William, Weddell, Alexander and Merrett, Geoff
(2019)
Efficient state retention through paged memory management for reactive transient computing.
Design Automation Conference, Las Vegas Convention Centre, Las Vegas, United States.
02 - 06 Jun 2019.
6 pp
.
(doi:10.1145/3316781.3317812).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Reactive transient computing systems preserve computational progress despite frequent power failures by suspending (saving state to nonvolatile memory) when detecting a power failure, and restoring once power returns. Existing methods inefficiently save and restore all allocated memory. We propose lightweight memory management that applies the concept of paging to load pages only when needed, and save only modified pages. We then develop a model that maximises available execution time by dynamically adjusting the suspend and restore voltage thresholds. Experiments on an MSP430FR5994 microcontroller show that our method reduces state retention overheads by up to 86.9% and executes algorithms up to 5.3 times faster than the state-of-the-art.
Text
dac-authorversion
- Accepted Manuscript
More information
Published date: 6 June 2019
Venue - Dates:
Design Automation Conference, Las Vegas Convention Centre, Las Vegas, United States, 2019-06-02 - 2019-06-06
Keywords:
Transient Computing, Internet of Things, Batteryless Computing
Identifiers
Local EPrints ID: 429666
URI: http://eprints.soton.ac.uk/id/eprint/429666
PURE UUID: 003ea0d8-e1d0-4c87-b824-d2f9cf71be6b
Catalogue record
Date deposited: 03 Apr 2019 16:30
Last modified: 16 Mar 2024 07:44
Export record
Altmetrics
Contributors
Author:
Sivert T. Sliper
Author:
Domenico Balsamo
Author:
Nikos Nikoleris
Author:
William Wang
Author:
Alexander Weddell
Author:
Geoff Merrett
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