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
크기와 이슈 폭이 증가함에 따라 명령어 이슈 큐는 오늘날의 슈퍼스칼라 마이크로프로세서에서 가장 에너지를 많이 소비하는 장치 중 하나가 되었습니다. 이 문서에서는 명령 스트림의 변화하는 계산 요구 사항에 맞게 런타임 시 기능이 동적으로 조정되는 적응형 문제 대기열의 에너지 손실을 줄이기 위한 새로운 아키텍처 기술을 제시합니다. 기존 방식과 달리 이 기술은 큐 설계의 새로운 자유, 즉 액세스당 전압을 활용합니다. 적응형 큐에서 작동하는 로딩 커패시턴스는 시간에 따라 달라지므로 클록 주기 예산이 비효율적으로 활용됩니다. 우리는 사용되지 않는 사이클 시간을 공급 전압과 절충하여 큐 기능이 감소할 때 전압 레벨을 낮추고 큐의 리소스 활성화에 따라 전압 레벨을 높이는 것을 제안합니다. 실험에 따르면 이 접근 방식은 큰 성능 및 영역 오버헤드 없이 문제 대기열 에너지를 최대 39%까지 절약할 수 있는 것으로 나타났습니다.
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.
부
Vasily G. MOSHNYAGA, "Issue Queue Energy Reduction through Dynamic Voltage Scaling" in IEICE TRANSACTIONS on Electronics,
vol. E85-C, no. 2, pp. 272-278, February 2002, doi: .
Abstract: With increased size and issue-width, instruction issue queue becomes one of the most energy consuming units in today's superscalar microprocessors. This paper presents a novel architectural technique to reduce energy dissipation of adaptive issue queue, whose functionality is dynamically adjusted at runtime to match the changing computational demands of instruction stream. In contrast to existing schemes, the technique exploits a new freedom in queue design, namely the voltage per access. Since loading capacitance operated in the adaptive queue varies in time, the clock cycle budget becomes inefficiently exploited. We propose to trade-off the unused cycle time with supply voltage, lowering the voltage level when the queue functionality is reduced and increasing it with the activation of resources in the queue. Experiments show that the approach can save up to 39% of the issue queue energy without large performance and area overhead.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/e85-c_2_272/_p
부
@ARTICLE{e85-c_2_272,
author={Vasily G. MOSHNYAGA, },
journal={IEICE TRANSACTIONS on Electronics},
title={Issue Queue Energy Reduction through Dynamic Voltage Scaling},
year={2002},
volume={E85-C},
number={2},
pages={272-278},
abstract={With increased size and issue-width, instruction issue queue becomes one of the most energy consuming units in today's superscalar microprocessors. This paper presents a novel architectural technique to reduce energy dissipation of adaptive issue queue, whose functionality is dynamically adjusted at runtime to match the changing computational demands of instruction stream. In contrast to existing schemes, the technique exploits a new freedom in queue design, namely the voltage per access. Since loading capacitance operated in the adaptive queue varies in time, the clock cycle budget becomes inefficiently exploited. We propose to trade-off the unused cycle time with supply voltage, lowering the voltage level when the queue functionality is reduced and increasing it with the activation of resources in the queue. Experiments show that the approach can save up to 39% of the issue queue energy without large performance and area overhead.},
keywords={},
doi={},
ISSN={},
month={February},}
부
TY - JOUR
TI - Issue Queue Energy Reduction through Dynamic Voltage Scaling
T2 - IEICE TRANSACTIONS on Electronics
SP - 272
EP - 278
AU - Vasily G. MOSHNYAGA
PY - 2002
DO -
JO - IEICE TRANSACTIONS on Electronics
SN -
VL - E85-C
IS - 2
JA - IEICE TRANSACTIONS on Electronics
Y1 - February 2002
AB - With increased size and issue-width, instruction issue queue becomes one of the most energy consuming units in today's superscalar microprocessors. This paper presents a novel architectural technique to reduce energy dissipation of adaptive issue queue, whose functionality is dynamically adjusted at runtime to match the changing computational demands of instruction stream. In contrast to existing schemes, the technique exploits a new freedom in queue design, namely the voltage per access. Since loading capacitance operated in the adaptive queue varies in time, the clock cycle budget becomes inefficiently exploited. We propose to trade-off the unused cycle time with supply voltage, lowering the voltage level when the queue functionality is reduced and increasing it with the activation of resources in the queue. Experiments show that the approach can save up to 39% of the issue queue energy without large performance and area overhead.
ER -