An Efficient Virtual Channel Power-Gating Method using Dynamic Bypass in NoC Routers

With transistor sizes continuing to reduce, static power consumption has gradually become a major part of the overall power consumption of Network-on-chip (NoC). Therefore, many power gating methods have been proposed to reduce the increasing static power consumption. However, conventional power gating techniques introduce new problems such as high latency, low scalability and network disconnection. To address these issues, in this paper, we propose an efficient virtual channel (VC) power gating method using dynamic bypass. Firstly, each port of the router dynamically powers on and off the VC buffers based on network traffic, thus reducing the static power consumption of the router. Secondly, a straight bypass is created on each port to handle straight packets and a dynamic bypass shared by all ports is used to handle other types of packets. The bypass mechanism facilitates packet ejection, transmission, and injection operations even when the VC buffer is deactivated, effectively reducing packet latency and enhancing router performance through optimized data flow management. Finally, the experimental results show that the proposed power gating method reduces the static power consumption by 84.5% and packet latency by 8.4% on average compared to the baseline router, while increasing the area overhead by 13.29%.

Authors

Institutions

Publication Details

Journal
ACM Journal on Emerging Technologies in Computing Systems
Published
2026-09-30
DOI
https://doi.org/10.1145/3842736
Primary Topic
Interconnection Networks and Systems
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

An Efficient Virtual Channel Power-Gating Method using Dynamic Bypass in NoC Routers

Jianhua Li, Yiming Ouyang, Huaguo Liang, Jingyu Chen
ACM Journal on Emerging Technologies in Computing Systems
Interconnection Networks and Systems
article

An Efficient Virtual Channel Power-Gating Method using Dynamic Bypass in NoC Routers

Jianhua Li, Yiming Ouyang, Huaguo Liang, Jingyu Chen
article en

Abstract

With transistor sizes continuing to reduce, static power consumption has gradually become a major part of the overall power consumption of Network-on-chip (NoC). Therefore, many power gating methods have been proposed to reduce the increasing static power consumption. However, conventional power gating techniques introduce new problems such as high latency, low scalability and network disconnection. To address these issues, in this paper, we propose an efficient virtual channel (VC) power gating method using dynamic bypass. Firstly, each port of the router dynamically powers on and off the VC buffers based on network traffic, thus reducing the static power consumption of the router. Secondly, a straight bypass is created on each port to handle straight packets and a dynamic bypass shared by all ports is used to handle other types of packets. The bypass mechanism facilitates packet ejection, transmission, and injection operations even when the VC buffer is deactivated, effectively reducing packet latency and enhancing router performance through optimized data flow management. Finally, the experimental results show that the proposed power gating method reduces the static power consumption by 84.5% and packet latency by 8.4% on average compared to the baseline router, while increasing the area overhead by 13.29%.

ACM Journal on Emerging Technologies in Computing Systems
Hefei University of Technology (CN)
Affordable and clean energy
Openalex Percentile: Top 9%
Interconnection Networks and Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.