Contribution to a conference proceedings

FZJ-2024-06447

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Performance and Power: Systematic Evaluation of AI Workloads on Accelerators with CARAML

John, C. M.FZJ* ; Herten, A. (Corresponding author)FZJ* ; Penke, C. (Corresponding author)FZJ* ; Nassyr, S. (Corresponding author)FZJ*

2024

This record in other databases:    

Please use a persistent id in citations: doi:10.1109/SCW63240.2024.00158  doi:10.34734/FZJ-2024-06447

Abstract: The rapid advancement of machine learning (ML) technologies has driven the development of specialized hardware accelerators designed to facilitate more efficient model training. This paper introduces the CARAML benchmark suite, which is employed to assess performance and energy consumption during the training of transformer-based large language models and computer vision models on a range of hardware accelerators, including systems from NVIDIA, AMD, and Graphcore. CARAML provides a compact, automated, extensible, and reproducible framework for assessing the performance and energy of ML workloads across various novel hardware architectures. The design and implementation of CARAML, along with a custom power measurement tool called jpwr, are discussed in detail.

Note: Also available at: https://arxiv.org/abs/2409.12994

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511) (POF4-511)
2. 5121 - Supercomputing & Big Data Facilities (POF4-512) (POF4-512)
3. OpenGPT-X - Aufbau eines Gaia-X Knotens für große KI-Sprachmodelle und innovative Sprachapplikations-Services; Teilvorhaben: Optimierung und Skalierung auf großen HPC-Systemen (68GX21007F) (68GX21007F)
4. MAELSTROM - MAchinE Learning for Scalable meTeoROlogy and cliMate (955513) (955513)
5. Verbundprojekt: MAELSTROM - Skalierbarkeit von Anwendungen des Maschinellen Lernens in den Bereichen Wetter und Klimawissenschaften für das zukünftige Supercomputing (16HPC029) (16HPC029)
6. ATML-X-DEV - ATML Accelerating Devices (ATML-X-DEV) (ATML-X-DEV)

Appears in the scientific report 2024

---

Database coverage:

---