Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Comprehensive Analysis of Process Energy Consumption on Multi-Socket Systems with GPUs

Luis G. León-Vega, Niccolò Tosato, Stefano Cozzini

TL;DR

This work investigates the impact of executed instructions on overall power consumption, providing insights into the comprehensive behaviour of HPC systems and introduces two novel mathematical models to estimate a process's energy consumption based on the total node energy, process usage, and a normalised vector of the probability distribution of instruction types for CPU and GPU processes.

Abstract

Robustly estimating energy consumption in High-Performance Computing (HPC) is essential for assessing the energy footprint of modern workloads, particularly in fields such as Artificial Intelligence (AI) research, development, and deployment. The extensive use of supercomputers for AI training has heightened concerns about energy consumption and carbon emissions. Existing energy estimation tools often assume exclusive use of computing nodes, a premise that becomes problematic with the advent of supercomputers integrating microservices, as seen in initiatives like Acceleration as a Service (XaaS) and cloud computing. This work investigates the impact of executed instructions on overall power consumption, providing insights into the comprehensive behaviour of HPC systems. We introduce two novel mathematical models to estimate a process's energy consumption based on the total node energy, process usage, and a normalised vector of the probability distribution of instruction types for CPU and GPU processes. Our approach enables energy accounting for specific processes without the need for isolation. Our models demonstrate high accuracy, predicting CPU power consumption with a mere 1.9% error. For GPU predictions, the models achieve a central relative error of 9.7%, showing a clear tendency to fit the test data accurately. These results pave the way for new tools to measure and account for energy consumption in shared supercomputing environments.

A Comprehensive Analysis of Process Energy Consumption on Multi-Socket Systems with GPUs

TL;DR

This work investigates the impact of executed instructions on overall power consumption, providing insights into the comprehensive behaviour of HPC systems and introduces two novel mathematical models to estimate a process's energy consumption based on the total node energy, process usage, and a normalised vector of the probability distribution of instruction types for CPU and GPU processes.

Abstract

Robustly estimating energy consumption in High-Performance Computing (HPC) is essential for assessing the energy footprint of modern workloads, particularly in fields such as Artificial Intelligence (AI) research, development, and deployment. The extensive use of supercomputers for AI training has heightened concerns about energy consumption and carbon emissions. Existing energy estimation tools often assume exclusive use of computing nodes, a premise that becomes problematic with the advent of supercomputers integrating microservices, as seen in initiatives like Acceleration as a Service (XaaS) and cloud computing. This work investigates the impact of executed instructions on overall power consumption, providing insights into the comprehensive behaviour of HPC systems. We introduce two novel mathematical models to estimate a process's energy consumption based on the total node energy, process usage, and a normalised vector of the probability distribution of instruction types for CPU and GPU processes. Our approach enables energy accounting for specific processes without the need for isolation. Our models demonstrate high accuracy, predicting CPU power consumption with a mere 1.9% error. For GPU predictions, the models achieve a central relative error of 9.7%, showing a clear tendency to fit the test data accurately. These results pave the way for new tools to measure and account for energy consumption in shared supercomputing environments.
Paper Structure (9 sections, 21 equations, 3 figures, 3 tables)

This paper contains 9 sections, 21 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Process Power Consumption Analysis
  4. Energy Model for CPU workloads
  5. Energy Model for GPU workloads
  6. Model Experimental Results
  7. Conclusions and Future Work
  8. Acknowledgments.
  9. Disclosure of Interests.

Figures (3)

  • Figure 1: Experiments behaviour when scaling in degree of parallelism (System CPU Usage) and the instructions executed. On the left, the socket power consumption is reported as the usage scales, whereas on the right, the histogram of the instructions illustrates the experiment's nature.
  • Figure 2: Experiments behaviour when scaling the GPU occupation and the instructions. On the left, the GPU card power consumption is reported as the usage scales, whereas on the right, the histogram of the instructions illustrates the experiment's nature.
  • Figure 3: CPU and GPU model performance in power consumption prediction. On the left, the CPU model is fitted using non-negative linear regression. On the right, the GPU model is fitted using least-squares linear regression.