Energy Minimisation for Processor Cores using Variable Supply Voltages
Energy Minimisation for Processor Cores using Variable Supply Voltages
This paper presents an efficient technique for energy minimisation based on dynamically variable supply voltages. The efficiency is achieved by considering that power consumed by the core depends on the computation carried out on it. Energy is further minimised by allowing the voltage supply to vary continuously. The computational delay introduced due to supply voltage reduction is tolerated by exploiting the idle time while meeting the system time constraint. The energy minimisation problem is formulated and solved using a iterative improvement algorithm. Various examples with different task set complexities show the efficiency of the proposed technique.
10/1-10/4
Schmitz, Marcus T.
76acfda5-3d49-47a0-a38b-a73273f2ba21
Al-Hashimi, Bashir M.
0b29c671-a6d2-459c-af68-c4614dce3b5d
September 2000
Schmitz, Marcus T.
76acfda5-3d49-47a0-a38b-a73273f2ba21
Al-Hashimi, Bashir M.
0b29c671-a6d2-459c-af68-c4614dce3b5d
Schmitz, Marcus T. and Al-Hashimi, Bashir M.
(2000)
Energy Minimisation for Processor Cores using Variable Supply Voltages.
IEE System on a Chip Workshop.
.
Record type:
Conference or Workshop Item
(Other)
Abstract
This paper presents an efficient technique for energy minimisation based on dynamically variable supply voltages. The efficiency is achieved by considering that power consumed by the core depends on the computation carried out on it. Energy is further minimised by allowing the voltage supply to vary continuously. The computational delay introduced due to supply voltage reduction is tolerated by exploiting the idle time while meeting the system time constraint. The energy minimisation problem is formulated and solved using a iterative improvement algorithm. Various examples with different task set complexities show the efficiency of the proposed technique.
More information
Published date: September 2000
Venue - Dates:
IEE System on a Chip Workshop, 2000-09-01
Organisations:
Electronic & Software Systems
Identifiers
Local EPrints ID: 253758
URI: http://eprints.soton.ac.uk/id/eprint/253758
PURE UUID: 3ccd3390-84a7-44ed-a783-0d8852a53cab
Catalogue record
Date deposited: 26 Sep 2000
Last modified: 14 Mar 2024 05:28
Export record
Contributors
Author:
Marcus T. Schmitz
Author:
Bashir M. Al-Hashimi
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.
Loading...View more statistics
