CarbonScaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon-Efficiency
CarbonScaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon-Efficiency
复制标题
DOI：
10.1145/3626788
复制
发表时间：
2023-02
期刊：
影响因子：
--
通讯作者：
Walid A. Hanafy;Qianlin Liang;Noman Bashir;David E. Irwin;Prashant J. Shenoy
中科院分区：
文献类型：
--
作者：
Walid A. Hanafy;Qianlin Liang;Noman Bashir;David E. Irwin;Prashant J. Shenoy
云平台越来越重视可持续性，并减少其运营碳足迹。减少碳排放的一种常见方法是利用许多云工作负载固有的时间灵活性，在能源最环保的时期执行这些工作负载，并在其他时间暂停它们。由于这种暂停-恢复方法可能会导致作业完成时间的长延迟，因此我们提出了一种利用云中批处理工作负载的弹性来优化其碳排放的新方法。我们的方法基于“碳缩放”的概念，类似于云自动缩放，其中作业根据网格能源碳成本的波动动态地改变其服务器分配。基于众所周知的边际资源分配问题，我们开发了一种贪心算法，通过碳标度最小化工作的碳排放。我们在Kubernetes中实现了一个CarbonScaler原型，使用它的自动缩放功能和一个分析工具来指导在云中批量应用程序的碳效率部署。然后，我们在商业云平台上使用真实世界的机器学习训练和MPI作业来评估CarbonScaler，并表明它可以产生i)比碳不可知执行节省51%的碳；Ii)比最先进的暂停恢复政策高出37%；iii) 8超过最佳静态缩放策略。
Cloud platforms are increasing their emphasis on sustainability and reducing their operational carbon footprint. A common approach for reducing carbon emissions is to exploit the temporal flexibility inherent to many cloud workloads by executing them in periods with the greenest energy and suspending them at other times. Since such suspend-resume approaches can incur long delays in job completion times, we present a new approach that exploits the elasticity of batch workloads in the cloud to optimize their carbon emissions. Our approach is based on the notion of "carbon scaling," similar to cloud autoscaling, where a job dynamically varies its server allocation based on fluctuations in the carbon cost of the grid's energy. We develop a greedy algorithm for minimizing a job's carbon emissions via carbon scaling that is based on the well-known problem of marginal resource allocation. We implement a CarbonScaler prototype in Kubernetes using its autoscaling capabilities and an analytic tool to guide the carbon-efficient deployment of batch applications in the cloud. We then evaluate CarbonScaler using real-world machine learning training and MPI jobs on a commercial cloud platform and show that it can yield i) 51% carbon savings over carbon-agnostic execution; ii) 37% over a state-of-the-art suspend-resume policy; and iii) 8 over the best static scaling policy.