A Study of Operational Impact on Power Usage Effectiveness using Facility Metrics and Server Operation Logs in the K Computer

2020 IEEE International Conference on Cluster Computing (CLUSTER)(2020)

引用 2|浏览7
暂无评分
摘要
The official service period of the K computer ended in 2019. Most of the equipment has been enhanced and are being used as part of the infrastructure of the Fugaku successor system. Understanding the historical facility behavior of the K computer is valuable to ensure stable and energy-efficient operations over the next decade. The K computer was powered by electricity purchased from a utility company and energy generated by gas turbine power generators on the premises. To evaluate the energy efficiency of the entire center, we use a modified power usage effectiveness (PUE) metric that considers different forms of energy source purchased from utility companies and provides the value of the metric in the service period. To analyze the effect of operational impact on PUE, we use both the facility metrics and the server operation metrics extracted from the K computer logs. Applying the metric data to three case studies reveals that some maintenance operations degrade PUE. In particular, compared to emergency operations, annual maintenance operations tend to affect the PUE metric. Finally, we provide preliminary results that indicate an operational issue regarding the gas co-generation system.
更多
查看译文
关键词
power usage effectiveness,energy efficiency,data analysis,operations,co-generation system,high-performance computing,K computer
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要