一句话：提出基于机器学习的 ESAVE，利用少量硬件属性非侵入式预测服务器与虚拟机在不同利用率下的能耗。
随着数据中心和软件系统对基础设施的能耗不断增长，可持续软件工程受到广泛关注。为了在不侵入系统的前提下做出可持续的软件部署决策，需要准确刻画服务器或虚拟机的能耗。本文提出 ESAVE，一种基于机器学习的方法，利用少量硬件属性来表征服务器或虚拟机在不同利用率水平下的能耗。该方法通过对多种机器学习方法进行广泛探索，重点关注所需属性的最小化，同时保持良好精度。早期验证表明，尽管 ESAVE 是非侵入式的，其平均预测误差仅约 12%。
Sustainable software engineering has received a lot of attention in recent times, as we witness an ever-growing slice of energy use, for example, at data centers, as software systems utilize the underlying infrastructure. Characterizing servers for their energy use accurately without being intrusive, is therefore important to make sustainable software deployment choices. In this paper, we introduce ESAVE which is a machine learning-based approach that leverages a small set of hardware attributes to characterize a server or virtual machine’s energy usage across different levels of utilization. This is based upon an extensive exploration of multiple ML approaches, with a focus on a minimal set of required attributes, while showcasing good accuracy. Early validations show that ESAVE has only around 12% average prediction error, despite being non-intrusive.
@inproceedings{paperbot7309,
title = {ESAVE: Estimating Server and Virtual Machine Energy},
author = {Priyavanshi Pathania and Rohit Mehra and Vibhu Saujanya Sharma and Vikrant Kaulgud and Sanjay Podder and Adam P. Burden},
booktitle = {ASE},
year = {2022},
doi = {10.1145/3551349.3561170}
}