一句话：提出预测式而非反应式的细粒度 DVFS 机制，以更高效适应 GPU 工作负载的快速波动。
随着片上集成电压调节器（IVR）和快速自适应频率控制的发展，DVFS 的切换时间已从微秒级缩短到纳秒级，为提高能效带来巨大机会。然而，传统反应式 DVFS 难以跟上 GPU 工作负载需求的快速波动；本文因此提出“预测而非反应”的细粒度 DVFS 机制，通过预测工作负载需求来提前调整电压/频率，以更充分利用 V/f 电路技术。摘要未给出具体方法细节与量化评估结果，核心贡献在于提出这一预测式调控方向。
With the continuous improvement of on-chip integrated voltage regulators (IVRs) and fast, adaptive frequency control, dynamic voltage-frequency scaling (DVFS) transition times have shrunk from the microsecond to the nanosecond regime, providing immense opportunity to improve energy efficiency. The key to unlocking the continued improvement in V/f circuit technology is the creation of new, smarter DVFS mechanisms that better adapt to rapid fluctuations in workload demand.
@inproceedings{paperbot10122,
title = {Predict; Don't React for Enabling Efficient Fine-Grain DVFS in GPUs},
author = {Srikant Bharadwaj and Shomit Das and Kaushik Mazumdar and Bradford M. Beckmann and Stephen Kosonocky},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3623278.3624756}
}