Benchmark-based Study of CPU/GPU Power-Related Features through JAX and TensorFlow

TL;DR

This study evaluates power management techniques on CPUs using benchmarks with TensorFlow and JAX, revealing frequency limitation as the most effective method to optimize energy-delay trade-offs across different hardware and frameworks.

Contribution

It provides an empirical comparison of power management strategies across hardware and software frameworks, highlighting the effectiveness of frequency limitation in energy efficiency.

Findings

Frequency limitation improves Energy-Delay Product (EDP) significantly.

Running at maximum frequency can reduce EDP by a factor of 10.

Different frameworks (TensorFlow and JAX) respond differently to power management settings.

Abstract

Power management has become a crucial focus in the modern computing landscape, considering that {\em energy} is increasingly recognized as a critical resource. This increased the importance of all topics related to {\em energy-aware computing}. This paper presents an experimental study of three prevalent power management techniques that are {\em power limitation, frequency limitation}, and {\em ACPI/P-State governor modes} (OS states related to power consumption). Through a benchmark approach with a set of six computing kernels, we investigate {\em power/performance} trade-off with various hardware units and software frameworks (mainly TensorFlow and JAX). Our experimental results show that {\em frequency limitation} is the most effective technique to improve {\em Energy-Delay Product (EDP)}, which is a convolution of energy and running time. We also observe that running at the highest frequency compared to a reduced one could lead to a reduction of factor $\frac{1}{10}$ in EDP. Another noticeable fact is that frequency management shows a consistent behavior with different CPUs, whereas opposite effects sometimes occur between TensorFlow (TF) and JAX with the same power management settings.