# Verified Instruction-Level Energy Consumption Measurement for NVIDIA GPUs

arXiv: https://arxiv.org/abs/2002.07795 · HTML: https://mlsanity.com/paper/2002.07795/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Feb 2020.

## Built on by (most cited first, 10 of 10)

- [Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training](https://mlsanity.com/paper/2208.06102/) (2022)
- [Fast-SNN: Fast Spiking Neural Network by Converting Quantized ANN](https://mlsanity.com/paper/2305.19868/) (2023)
- [The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimization](https://mlsanity.com/paper/2505.06371/) (2025)
- [The Price of Prompting: Profiling Energy Use in Large Language Models Inference](https://mlsanity.com/paper/2407.16893/) (2024)
- [Dynamic GPU Energy Optimization for Machine Learning Training Workloads](https://mlsanity.com/paper/2201.01684/) (2022)
- [PolyThrottle: Energy-efficient Neural Network Inference on Edge Devices](https://mlsanity.com/paper/2310.19991/) (2023)
- [Dissecting Transformers: A CLEAR Perspective towards Green AI](https://mlsanity.com/paper/2510.02810/) (2025)
- [Energy awareness in low precision neural networks](https://mlsanity.com/paper/2202.02783/) (2022)
- [Who Wins the Race? (R Vs Python) - An Exploratory Study on Energy Consumption of Machine Learning Algorithms](https://mlsanity.com/paper/2508.17344/) (2025)
- [The Language-Energy Divide: Measuring Energy Costs of Multilingual LLM Inference](https://mlsanity.com/paper/2606.21869/) (2026)

## Abstract (arXiv)

GPUs are prevalent in modern computing systems at all scales. They consume a significant fraction of the energy in these systems. However, vendors do not publish the actual cost of the power/energy overhead of their internal microarchitecture. In this paper, we accurately measure the energy consumption of various PTX instructions found in modern NVIDIA GPUs. We provide an exhaustive comparison of more than 40 instructions for four high-end NVIDIA GPUs from four different generations (Maxwell, Pascal, Volta, and Turing). Furthermore, we show the effect of the CUDA compiler optimizations on the energy consumption of each instruction. We use three different software techniques to read the GPU on-chip power sensors, which use NVIDIA's NVML API and provide an in-depth comparison between these techniques. Additionally, we verified the software measurement techniques against a custom-designed hardware power measurement. The results show that Volta GPUs have the best energy efficiency of all the other generations for the different categories of the instructions. This work should aid in understanding NVIDIA GPUs' microarchitecture. It should also make energy measurements of any GPU kernel both efficient and accurate.