Low overhead performance monitoring for shared infrastructures
作者：
Highlights：
• This work analyzes the influence of monitoring overhead in virtualized environments.
• Results revealed a significant resource contention introduced by monitoring activity.
• Choosing only relevant counters reduces monitoring overhead and monetary costs.
• The proposed approach reduces the data dimensionality by means of feature selection.
• With our approach, the overhead can be one third lower than that usually observed.
摘要
•This work analyzes the influence of monitoring overhead in virtualized environments.•Results revealed a significant resource contention introduced by monitoring activity.•Choosing only relevant counters reduces monitoring overhead and monetary costs.•The proposed approach reduces the data dimensionality by means of feature selection.•With our approach, the overhead can be one third lower than that usually observed.
论文关键词：System monitoring,Performance counters,Performance optimization,Virtualization,Feature selection,Data science
论文评审过程：Received 20 August 2020, Revised 4 December 2020, Accepted 30 December 2020, Available online 5 January 2021, Version of Record 20 January 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.114558
