NVSentinel
NVIDIA Introduces Open Source Tools for Kubernetes GPU Cluster Operations
NVIDIA has detailed NVIDIA DSX OS, an open source software layer for configuring, certifying and maintaining accelerated Kubernetes infrastructure. The platform combines…
Together AI Details Auto Repair for GPU Clusters, From Fault to Healthy Node
Together AI has published a detailed look at how auto repair works in its GPU Clusters, describing a pipeline that goes from fault detection to a re-validated machine.…