An analytical methodology to derive power models based on hardware and software metrics

The use of models to predict the power con- sumption of a system is an appealing alternative to wattmeters since they avoid hardware costs and are easy to deploy. In this paper, we present a systematic methodology to build models with a reduced number of features in order to estimate power consumption at node level. We aim at building simple power models by performing a per-component analysis (CPU, mem- ory, network, I/O) through the execution of four stan- dard benchmarks. While they are executed, we collect information from all the available hardware counters and resource utilization metrics provided by the sys- tem. Based on correlations among the recorded metrics and their correlation with the instantaneous power, our methodology allows i) to identify the significant met- rics; and ii) to assign weights to the selected metrics in order to derive reduced models. The reduction also aims at extracting models that are based on a set of hardware counters and utilization metrics that can be obtained simultaneously and, thus, can be gathered and com- puted on-line. The utility of our procedure is validated using real-life applications on an Intel Sandy Bridge architecture