Coca4ai: checking energy behaviors on AI data centers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gay, Paul, Bilinski, Éric, Ligozat, Anne-Laure
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916331876515840
author Gay, Paul
Bilinski, Éric
Ligozat, Anne-Laure
author_facet Gay, Paul
Bilinski, Éric
Ligozat, Anne-Laure
contents Monitoring energy behaviors in AI data centers is crucial, both to reduce their energy consumption and to raise awareness among their users which are key actors in the AI field. This paper shows a proof of concept of easy and lightweight monitoring of energy behaviors at the scale of a whole data center, a user or a job submission. Our system uses software wattmeters and we validate our setup with per node accurate external wattmeters. Results show that there is an interesting potential from the efficiency point of view, providing arguments to create user engagement thanks to energy monitoring.
format Preprint
id arxiv_https___arxiv_org_abs_2407_15670
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Coca4ai: checking energy behaviors on AI data centers
Gay, Paul
Bilinski, Éric
Ligozat, Anne-Laure
Computers and Society
Monitoring energy behaviors in AI data centers is crucial, both to reduce their energy consumption and to raise awareness among their users which are key actors in the AI field. This paper shows a proof of concept of easy and lightweight monitoring of energy behaviors at the scale of a whole data center, a user or a job submission. Our system uses software wattmeters and we validate our setup with per node accurate external wattmeters. Results show that there is an interesting potential from the efficiency point of view, providing arguments to create user engagement thanks to energy monitoring.
title Coca4ai: checking energy behaviors on AI data centers
topic Computers and Society
url https://arxiv.org/abs/2407.15670