Datacenter Energy Optimized Power Profiles

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Narayanaswamy, Sreedhar, Patel, Pratikkumar Dilipkumar, Karlin, Ian, Gupta, Apoorv, Saripalli, Sudhir, Guo, Janey
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911303233175552
author Narayanaswamy, Sreedhar
Patel, Pratikkumar Dilipkumar
Karlin, Ian
Gupta, Apoorv
Saripalli, Sudhir
Guo, Janey
author_facet Narayanaswamy, Sreedhar
Patel, Pratikkumar Dilipkumar
Karlin, Ian
Gupta, Apoorv
Saripalli, Sudhir
Guo, Janey
contents This paper presents datacenter power profiles, a new NVIDIA software feature released with Blackwell B200, aimed at improving energy efficiency and/or performance. The initial feature provides coarse-grain user control for HPC and AI workloads leveraging hardware and software innovations for intelligent power management and domain knowledge of HPC and AI workloads. The resulting workload-aware optimization recipes maximize computational throughput while operating within strict facility power constraints. The phase-1 Blackwell implementation achieves up to 15% energy savings while maintaining performance levels above 97% for critical applications, enabling an overall throughput increase of up to 13% in a power-constrained facility. KEYWORDS GPU power management, energy efficiency, power profile, HPC optimization, Max-Q, Blackwell architecture
format Preprint
id arxiv_https___arxiv_org_abs_2510_03872
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Datacenter Energy Optimized Power Profiles
Narayanaswamy, Sreedhar
Patel, Pratikkumar Dilipkumar
Karlin, Ian
Gupta, Apoorv
Saripalli, Sudhir
Guo, Janey
Distributed, Parallel, and Cluster Computing
This paper presents datacenter power profiles, a new NVIDIA software feature released with Blackwell B200, aimed at improving energy efficiency and/or performance. The initial feature provides coarse-grain user control for HPC and AI workloads leveraging hardware and software innovations for intelligent power management and domain knowledge of HPC and AI workloads. The resulting workload-aware optimization recipes maximize computational throughput while operating within strict facility power constraints. The phase-1 Blackwell implementation achieves up to 15% energy savings while maintaining performance levels above 97% for critical applications, enabling an overall throughput increase of up to 13% in a power-constrained facility. KEYWORDS GPU power management, energy efficiency, power profile, HPC optimization, Max-Q, Blackwell architecture
title Datacenter Energy Optimized Power Profiles
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2510.03872