Skip to content
nodeza.learn
RoCE course / 4. Operations / Checkpoint

Module 4 checkpoint

Run a RoCE fabric in production: know where every counter lives, work a symptom to a cause in a fixed order, configure the switch half on Cumulus and Dell SONiC, and plumb RDMA into Kubernetes.

Module 4 of 5 in RoCE and RDMA for AI fabrics13 questionspass mark 80%about 12 minbelow the mark, retry after 24 h

Loading your checkpoint state…

What it covers

  1. 4.1The counter map: sysfs, ethtool and rdma statistic
  2. 4.2Triage in order: symptom to counter to cause
  3. 4.3MTU, RoCE LAG, and what has to be installed
  4. 4.4The switch half: Cumulus NVUE and Dell Enterprise SONiC
  5. 4.5RDMA in Kubernetes: the NVIDIA Network Operator