project

NVIDIA Cluster Readiness Engine

Open source Kubernetes controller from NVIDIA that certifies GPU cluster readiness by running distributed test workloads across node groups and reporting per-node results through custom resources.

Current clusters

build1 publisher

NVIDIA's NVCRE runs the NCCL test that Slurm gets from a single srun command

NVIDIA's open source Cluster Readiness Engine runs real distributed workloads on named node groups and reports which nodes failed. The published sample takes 42 minutes to fail an eight-node all-reduce.

Publishers:developer.nvidia.com

Reality

Evidence52
Adoption: Insufficient
Hype gap+28
Incentives84
Confidence56