Saved in:
Bibliographic Details
Main Authors: Zheng, Yusheng, Mao, Wenan, Cheng, Shuyi, Feng, Fuqiu, Li, Guangshui, Liao, Zhaoyan, Huang, Yongzhuo, Xiao, Zhenwei, Li, Yuqing, Quinn, Andi, Ma, Tao
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.29235
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Performance diagnosis in production-scale AI training is challenging because subtle OS-level issues can trigger cascading GPU delays and network slowdowns, degrading training efficiency across thousands of GPUs. Existing profiling tools are limited to single system layers, incur prohibitive overhead (10--30%), or lack continuous deployment capabilities, resulting in manual analyses spanning days. We argue that continuous, cross-layer observability enabled by OS-level instrumentation and layered differential diagnosis is necessary to address this gap. We introduce SysOM-AI, a production observability system that continuously integrates CPU stack profiling, GPU kernel tracing, and NCCL event instrumentation via adaptive hybrid stack unwinding and eBPF-based tracing, incurring less than 0.4% overhead. Deployed at Alibaba across over 80,000 GPUs for more than one year, SysOM-AI helped diagnose 94 confirmed production issues, reducing median diagnosis time from days to approximately 10 minutes.