Dissecting the software-based measurement of CPU energy consumption: a comparative analysis
Abstract
Every day, we experience the effects of the global warming: extreme weather events, major forest fires, storms, global warming, etc. The scientific community acknowledges that this crisis is a consequence of human activities where Information and Communications Technologies (ICT) are an increasingly important contributor. Computer scientists need tools for measuring the footprint of the code they produce. Running Average Power Limit (RAPL) is a low-level interface designed by Intel that provides a measure of the energy consumption of a CPU (and more) without the need for additional hardware. Since 2017, it is available on most computing devices, including non-Intel devices such as AMD processors. More and more people are using RAPL for energy measurement, mostly like a black box without deep knowledge of its behaviour. In this paper, we propose to come back to the basic mechanisms that allow to use RAPL measurements and present a critical analysis of their operations. For each mechanism, we release a reference implementation in Rust that avoids the pitfalls we detected in existing tools, improving correctness, timing accuracy and performance. In addition to long-established methods, we explore the suitability of the recent eBPF technology for working with RAPL. We also provide an experimental study with multiple benchmarks and processor models in order to evaluate the efficiency of the various mechanisms and their impact on parallel software. Our experiments show that no mechanism provides a significant performance advantage over the others. However, they differ significantly in terms of ease-of-use and resiliency. We believe that this work will help the community to develop correct, resilient and lightweight measurement tools, based on the mechanism that suits their needs.
Origin : Files produced by the author(s)
