Server Energy Consumption Model Based on ConvLSTM in Mobile Edge Computing
李小龙 1李曦 2杨凌峰 1黄华1
扫码查看
点击上方二维码区域,可以放大扫码查看
作者信息
1. 湖南工商大学计算机学院,湖南长沙 410205
2. 湖南工商大学前沿交叉学院,湖南长沙 410205
折叠
摘要
针对现有能耗模型对动态工作负载波动具有低敏感性和低精度的问题,该文基于卷积长短期记忆(convolutional long short-term memory,ConvLSTM)神经网络,提出了用于移动边缘计算的服务器智能能耗模型(intelligence server energy consumption model,IECM),用于预测和优化服务器的能量消耗.通过收集服务器运行时间参数,使用熵值法筛选和保留显著影响服务器能耗的参数.基于选定的参数,利用ConvLSTM神经网络训练服务器能耗模型的深度网络.与现有的能耗模型相比,IECM在CPU密集型、I/O密集型、内存密集型和混合型任务上,能够适应服务器工作负载的动态变化,并在能耗预测上具有更好的准确性.
Abstract
To address the issue of low sensitivity and accuracy of existing energy con-sumption models in accommodating dynamic workload fluctuations,this paper proposes an intelligence server energy consumption model(IECM)based on the convolutional long short-term memory(ConvLSTM)neural network in mobile edge computing,which is used to predict and optimize energy consumption in servers.By collecting server runtime param-eters and using the entropy method to filter and retain parameters significantly affecting server energy consumption,a deep network for training the server energy consumption model is constructed based on the selected parameters using the ConvLSTM neural net-work.Compared with existing energy consumption models,IECM exhibits superior adapt-ability to dynamic changes in server workload in CPU-intensive,I/O-intensive,memory-intensive,and mixed tasks,offering enhanced accuracy in energy consumption prediction.
关键词
卷积长短期记忆/能耗预测/智能功率模型/功率建模
Key words
convolutional long short-term memory(ConvLSTM)/energy consumption prediction/intelligence power model/power modeling