June 22, 2026 ARGUS: Production-Scale Tracing and Performance Diagnosis for over 10,000-GPU Clusters 针对万卡 GPU 集群的低开销、细粒度、常开追踪与实时分析系统,总开销 <2%,内核事件压缩 3,700x gpu-cluster performance-diagnosis tracing distributed-training llm-training observability