Performance and Power: Systematic Evaluation of AI Workloads on Accelerators with CARAML
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913734112313344 |
|---|---|
| author | John, Chelsea Maria Nassyr, Stepan Penke, Carolin Herten, Andreas |
| author_facet | John, Chelsea Maria Nassyr, Stepan Penke, Carolin Herten, Andreas |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12994 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Performance and Power: Systematic Evaluation of AI Workloads on Accelerators with CARAML John, Chelsea Maria Nassyr, Stepan Penke, Carolin Herten, Andreas Hardware Architecture Artificial Intelligence Distributed, Parallel, and Cluster Computing Machine Learning Performance 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. |
| title | Performance and Power: Systematic Evaluation of AI Workloads on Accelerators with CARAML |
| topic | Hardware Architecture Artificial Intelligence Distributed, Parallel, and Cluster Computing Machine Learning Performance |
| url | https://arxiv.org/abs/2409.12994 |