Tag: #NVLink
Kubernetes, DevOps and MLOps, written from things we ran and measured ourselves · 2 posts
GPU Hardware and Driver Failure Diagnosis
This guide covers DCGM metrics, XID codes, ECC, NVLink, throttling, PCIe degradation, and how a single slow GPU slows down the entire workload based on NVIDIA official documentation, connecting them to Kubernetes isolati
2026-10-07 · 32 min read #DCGM#XID#ECC#NVLink#스로틀링GPU Networks Read via RDMA and NCCL: InfiniBand, RoCE, Kubernetes Deployment
We break down communication layers and bandwidth differences inside and outside GPU servers, explain the differences between InfiniBand and RoCE v2, GPUDirect RDMA, NCCL operation and measurement methods, and common fail
2026-10-07 · 41 min read #RDMA#InfiniBand#RoCE#NCCL#GPUDirect