SPARe: Stacked Parallelism with Adaptive Reordering for Fault-Tolerant LLM Pretraining Systems with 100k+ GPUs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Lee, Jin, Chen, Zhonghao, He, Xuhang, Underwood, Robert, Nicolae, Bogdan, Cappello, Franck, Lu, Xiaoyi, Di, Sheng, Zhang, Zheng
Format:	Preprint
Published:	2026
Subjects:	Distributed, Parallel, and Cluster Computing Systems and Control
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2603.00357