Reproduce “PowerSensor3: A Fast and Accurate Open Source Power Measurement Tool” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Steven van der Vlugt, Leon Oostrum, Gijs Schoonderbeek, Ben van Werkhoven, Bram Veenboer, Krijn Doekemeijer, John W. Romein

PowerSensor3 is an open-source, cost-effective power measurement tool that provides up to 20 kHz sampling rate for individual components like GPUs, FPGAs, and SSDs. It enables fine-grained energy profiling and accelerated auto-tuning processes, demonstrating a 3.25x speedup in optimizing GPU kernels and revealing power consumption patterns not observable with built-in sensors.

Key insights

  • Sub-millisecond power measurement granularity is essential for accurately profiling and optimizing rapidly executing code segments and understanding transient power behaviors in modern hardware.
  • Direct integration of high-resolution power measurements into auto-tuning frameworks can significantly reduce optimization time and identify energy-efficient configurations not discoverable by performance-centric metrics alone.
  • Performance metrics like bandwidth do not always correlate directly with power consumption (e.g., in SSDs during garbage collection), underscoring the necessity of external, precise power monitoring to fully understand device energy footprints.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.