NVIDIA details Attribution tool for diagnosing distributed AI failures
NVIDIA says its open-source Attribution system correlates workload evidence to identify root causes in distributed AI failures.
NVIDIA says automated recovery can reduce training downtime
NVIDIA described Recovery, part of DSX OS, as an open-source service to detect interruptions and restart distributed training jobs.
NVIDIA details Kubernetes recipe for Vera Rubin NVL72 support
NVIDIA says its AICR tool can help Kubernetes clusters validate and reproduce configurations for Vera Rubin NVL72 support.
NVIDIA details Workload Map for Kubernetes AI frameworks
NVIDIA says Workload Map provides a shared way to describe Kubernetes workloads across frameworks, with support for JobSet, KAI Scheduler and Run:ai.
NVIDIA details Kubernetes runtime recipe for Dynamo
NVIDIA outlined AICR recipes for deploying and validating Dynamo on Kubernetes, including routing options, pinned components and validation checks.
NVIDIA outlines GDS diagnostics and setup tools for GPU storage paths
NVIDIA details gds-diag, a toolset for checking GPUDirect Storage support, configuration and container compatibility across Linux environments.
KubeRay best practices focus on production scale and cluster failures
Anyscale outlines production KubeRay practices for logs, GPU health, autoscaler tuning and head-node resilience, with Connect in private preview.
AMD, Oracle detail Pensando DPU role in Acceleron SmartNICs
AMD says Pensando DPUs underpin Oracle Acceleron SmartNICs to offload networking, storage and security for AI and cloud workloads.
NVIDIA and Oracle expand Vera Rubin to multi-gigawatt OCI scale
NVIDIA and Oracle said OCI Superclusters will expand to Vera Rubin NVL72 at multi-gigawatt scale, with Vera CPUs now generally available on OCI.
NVIDIA Mission Control 2.5 extends autonomous recovery to Kubernetes
NVIDIA says Mission Control 2.5 brings autonomous recovery to Kubernetes workloads managed by Run:ai, adding job and hardware remediation controls.