At present, the major drawback for mobile phones is the issue of power consumption. As one of the alternatives to decrease the power consumption of standard, power-hungry location-based services usually require the kn...At present, the major drawback for mobile phones is the issue of power consumption. As one of the alternatives to decrease the power consumption of standard, power-hungry location-based services usually require the knowledge of how individual phone features consume power. A typical phone feature is that the applications related to multimedia streaming utilize more power while receiving, processing, and displaying the multimedia contents, thus contributing to the increased power consumption. There is a growing concern that current battery modules have limited capability in fulfilling the long-term energy need for the progress on the mobile phone because of increasing power consumption during multimedia streaming processes. Considering this, in this paper, we provide an offline meaning sleep-mode method to compute the minimum power consumption comparing with the power-on solution to save power by implementing energy rate adaptation(RA) mechanism based on mobile excess energy level purpose to save battery power use. Our simulation results show that our RA method preserves efficient power while achieving better throughput compared with the mechanism without rate adaptation(WRA).展开更多
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 consump...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。展开更多
基金supported by X-Project funded by the Ministry of Science,ICT&Future Planning under Grant No.NRF-2015R1A2A1A16074929
文摘At present, the major drawback for mobile phones is the issue of power consumption. As one of the alternatives to decrease the power consumption of standard, power-hungry location-based services usually require the knowledge of how individual phone features consume power. A typical phone feature is that the applications related to multimedia streaming utilize more power while receiving, processing, and displaying the multimedia contents, thus contributing to the increased power consumption. There is a growing concern that current battery modules have limited capability in fulfilling the long-term energy need for the progress on the mobile phone because of increasing power consumption during multimedia streaming processes. Considering this, in this paper, we provide an offline meaning sleep-mode method to compute the minimum power consumption comparing with the power-on solution to save power by implementing energy rate adaptation(RA) mechanism based on mobile excess energy level purpose to save battery power use. Our simulation results show that our RA method preserves efficient power while achieving better throughput compared with the mechanism without rate adaptation(WRA).
文摘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。
