The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
L'affectation de plusieurs tensions d'alimentation (MSV) est un moyen très efficace de réduire la consommation d'énergie. De nombreux algorithmes existants fonctionnent très bien en matière de réduction de puissance. Cependant, ils ne règlent pas le problème de zone des changeurs de niveau. Dans certains cas, bien que l'on obtienne un résultat supérieur en réduisant la consommation d'énergie, de nombreux leviers de niveau supplémentaires sont nécessaires pour que la surface du circuit dépasse les spécifications. Dans cet article, nous présentons une approche efficace d'affectation MSV basée sur la programmation linéaire entière (ILP) pour résoudre deux problèmes avec des objectifs différents. Pour l'objectif de réduction de puissance sous contrainte de timing, par rapport à l'algorithme GECVS, la consommation d'énergie obtenue par notre approche proposée peut être encore réduite de 0 à 5.46 % et le nombre de changeurs de niveau est amélioré de 16.31 % en moyenne. Pour l'objectif de réduction de puissance sous contraintes à la fois de timing et de zone des shifters de niveau, l'amélioration moyenne de la consommation d'énergie obtenue par notre algorithme est toujours meilleure que celle de GECVS tout en réduisant le nombre de shifters de niveau de 22.92% en moyenne. De plus, étant donné une contrainte de consommation totale d’énergie, notre algorithme générera une conception ayant un retard de circuit minimum. Les résultats expérimentaux montrent que l'algorithme d'affectation MSV basé sur ILP proposé résout différents problèmes de manière flexible.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Hsi-An CHIEN, Cheng-Chiang LIN, Hsin-Hsiung HUANG, Tsai-Ming HSIEH, "Optimal Supply Voltage Assignment under Timing, Power and Area Constraints" in IEICE TRANSACTIONS on Fundamentals,
vol. E93-A, no. 4, pp. 761-768, April 2010, doi: 10.1587/transfun.E93.A.761.
Abstract: Multiple supply voltage (MSV) assignment is a highly effective means of reducing power consumption. Many existing algorithms perform very well for power reduction. However, they do not handle the area issue of level shifters. In some cases, although one gets a superior result to reduce the power consumption, but many extra level shifters are needed to add so that the circuit area will be over the specification. In this paper, we present an effective integer linear programming (ILP)-based MSV assignment approach to solve two problems with different objectives. For the objective of power reduction under timing constraint, compared with GECVS algorithm, the power consumption obtained by our proposed approach can be further reduced 0 to 5.46% and the number of level shifters is improved 16.31% in average. For the objective of power reduction under constraints of both timing and area of level shifters, the average improvement of power consumption obtained by our algorithm is still better than GECVS while reducing the number of level shifters by 22.92% in average. In addition, given a constraint of total power consumption, our algorithm will generate a design having minimum circuit delay. Experimental results show that the proposed ILP-based MSV assignment algorithm solves different problems flexibly.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E93.A.761/_p
Copier
@ARTICLE{e93-a_4_761,
author={Hsi-An CHIEN, Cheng-Chiang LIN, Hsin-Hsiung HUANG, Tsai-Ming HSIEH, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Optimal Supply Voltage Assignment under Timing, Power and Area Constraints},
year={2010},
volume={E93-A},
number={4},
pages={761-768},
abstract={Multiple supply voltage (MSV) assignment is a highly effective means of reducing power consumption. Many existing algorithms perform very well for power reduction. However, they do not handle the area issue of level shifters. In some cases, although one gets a superior result to reduce the power consumption, but many extra level shifters are needed to add so that the circuit area will be over the specification. In this paper, we present an effective integer linear programming (ILP)-based MSV assignment approach to solve two problems with different objectives. For the objective of power reduction under timing constraint, compared with GECVS algorithm, the power consumption obtained by our proposed approach can be further reduced 0 to 5.46% and the number of level shifters is improved 16.31% in average. For the objective of power reduction under constraints of both timing and area of level shifters, the average improvement of power consumption obtained by our algorithm is still better than GECVS while reducing the number of level shifters by 22.92% in average. In addition, given a constraint of total power consumption, our algorithm will generate a design having minimum circuit delay. Experimental results show that the proposed ILP-based MSV assignment algorithm solves different problems flexibly.},
keywords={},
doi={10.1587/transfun.E93.A.761},
ISSN={1745-1337},
month={April},}
Copier
TY - JOUR
TI - Optimal Supply Voltage Assignment under Timing, Power and Area Constraints
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 761
EP - 768
AU - Hsi-An CHIEN
AU - Cheng-Chiang LIN
AU - Hsin-Hsiung HUANG
AU - Tsai-Ming HSIEH
PY - 2010
DO - 10.1587/transfun.E93.A.761
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E93-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2010
AB - Multiple supply voltage (MSV) assignment is a highly effective means of reducing power consumption. Many existing algorithms perform very well for power reduction. However, they do not handle the area issue of level shifters. In some cases, although one gets a superior result to reduce the power consumption, but many extra level shifters are needed to add so that the circuit area will be over the specification. In this paper, we present an effective integer linear programming (ILP)-based MSV assignment approach to solve two problems with different objectives. For the objective of power reduction under timing constraint, compared with GECVS algorithm, the power consumption obtained by our proposed approach can be further reduced 0 to 5.46% and the number of level shifters is improved 16.31% in average. For the objective of power reduction under constraints of both timing and area of level shifters, the average improvement of power consumption obtained by our algorithm is still better than GECVS while reducing the number of level shifters by 22.92% in average. In addition, given a constraint of total power consumption, our algorithm will generate a design having minimum circuit delay. Experimental results show that the proposed ILP-based MSV assignment algorithm solves different problems flexibly.
ER -