Redox: Improving I/O Efficiency of Model Training Through File Redirection
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909946970374144 |
|---|---|
| author | Li, Yuhao Shi, Xuanhua Zhao, Yunfei Zhou, Yongluan Hua, Yusheng Qian, Xuehai |
| author_facet | Li, Yuhao Shi, Xuanhua Zhao, Yunfei Zhou, Yongluan Hua, Yusheng Qian, Xuehai |
| contents | This paper proposes Redox, a training data management system designed to achieve high I/O efficiency. The key insight is a new observation of file redirection: for model training, when training data in one file is requested, the system has the flexibility to return the data of another file. Based on this property, Redox starts with a bold design principle that chunks of data files are always read from disk in batch, and once loaded, all files in the chunk will be consumed without being loaded again. We propose efficient local and distributed file read protocol based on this principle that both minimizes the wasted data read and enables opportunistic prefetch from remote node. Moreover, we analyze file redirection's impact on randomness, and show that it has little effects on training efficiency. Experimental results indicate that Redox significantly accelerates data fetching in training, achieving up to a 4.57x improvement in end-to-end training compared to PyTorch. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_16280 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Redox: Improving I/O Efficiency of Model Training Through File Redirection Li, Yuhao Shi, Xuanhua Zhao, Yunfei Zhou, Yongluan Hua, Yusheng Qian, Xuehai Distributed, Parallel, and Cluster Computing This paper proposes Redox, a training data management system designed to achieve high I/O efficiency. The key insight is a new observation of file redirection: for model training, when training data in one file is requested, the system has the flexibility to return the data of another file. Based on this property, Redox starts with a bold design principle that chunks of data files are always read from disk in batch, and once loaded, all files in the chunk will be consumed without being loaded again. We propose efficient local and distributed file read protocol based on this principle that both minimizes the wasted data read and enables opportunistic prefetch from remote node. Moreover, we analyze file redirection's impact on randomness, and show that it has little effects on training efficiency. Experimental results indicate that Redox significantly accelerates data fetching in training, achieving up to a 4.57x improvement in end-to-end training compared to PyTorch. |
| title | Redox: Improving I/O Efficiency of Model Training Through File Redirection |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2505.16280 |