NVIDIA unveils Minerva, an end-to-end AIOps observability platform for AI factories

NVIDIA has launched Minerva, an AIOps platform designed to help teams detect problems early and keep so-called "AI factories" productive. According to the announcement on the company's blog on October 22, 2026, the platform brings together telemetry from AI infrastructure and workloads and applies purpose-built AI models to turn that data into timely, actionable insights.

The approach starts from a premise: AI factories are among the most complex operations in existence, with huge numbers of GPUs working across high-speed networks to train models and serve applications. At that scale, a single degraded component can slow down an entire job. NVIDIA cites signals that can go unnoticed, such as a GPU running hotter than its peers or an NVMe drive losing performance under load.

The platform focuses on "AI factory yield," which NVIDIA defines as a measure of the useful work the infrastructure produces, in terms of GPU hours, goodput, cluster availability and token output.

A dynamic view of the factory. NVIDIA argues that traditional monitoring tools split the data center into separate views: networking, compute, site reliability (SRE) and machine learning teams each see a different part. Minerva connects those signals across NVIDIA GPUs, networking, storage, schedulers and workloads. From them it builds a dynamic knowledge graph that logically maps the infrastructure (GPUs, SuperNICs, switches and cables) and the jobs running on it. According to the company, this context makes it possible to trace a slow training job back to the component affecting it, even in a factory with hundreds of thousands of GPUs.

Analysis and correlation. NVIDIA says Minerva analyzes billions of data points per hour using supervised and unsupervised machine learning. It learns normal behavior, detects previously unseen anomalies, predicts emerging failures and correlates related events based on the factory's topology. Instead of sending every symptom to an operator, it can group an "event storm" into a single incident with a probable root cause.

Assistant and agent integration. The platform includes a native AI assistant and connects with external agentic AI tools, adding a natural-language interface grounded in live telemetry. Operators can investigate cluster health and receive remediation recommendations through conversation. Teams can keep a human in the loop to apply fixes manually, or connect Minerva to automation via infrastructure APIs and external systems.

Stated benefits. NVIDIA argues that grouping related problems into a single actionable item makes it possible to shift from reactive to proactive maintenance and maximize GPU uptime. It sums up the value in three metrics: longer time between interruptions (detecting problems before they occur so jobs run longer without incidents), shorter time to repair (pinpointing the problem when an incident occurs) and greater visibility (filtering the data stream down to the alerts that matter). It should be noted that these are the company's own claims; the article provides no specific figures or independent measurements.

Customers and partners. IREN is among the partners using Minerva. Its chief technology officer, Denis Skrinnikoff, said the platform gives the operations team a single view of the network, the GPUs and the jobs running on them, allowing it to focus more on proactive maintenance than on troubleshooting. According to the announcement, Aolani, Bitdeer AI, IBM, Jio, Lambda, Quake AI and Tesla also plan to deploy Minerva. In addition, Dell Technologies, Lenovo and Supermicro will offer Minerva as part of their AI factory operations portfolios.

Context. The announcement is part of NVIDIA's series of European news around GTC Berlin 2026. The text does not detail pricing, general availability or specific technical requirements, so those aspects remain to be clarified.