NVIDIA Minerva Platform Offers AI Factory Observability

NVIDIA has released Minerva, an artificial intelligence operations platform designed to monitor and maintain large-scale AI factories. The system aggregates telemetry data from infrastructure components and workloads, applying purpose-built models to generate actionable insights. This aims to help engineering teams identify issues earlier and sustain productivity across vast GPU clusters.

Root cause

Traditional monitoring tools often fragment data center visibility, separating networking, compute and machine learning views. Minerva addresses this by connecting signals across NVIDIA GPUs, storage, schedulers and networks. It builds a dynamic knowledge graph that maps the physical infrastructure and the jobs running on it. This context allows operators to trace performance issues to specific components, even in facilities with hundreds of thousands of GPUs.

Attack path

The platform processes billions of data points per hour using supervised and unsupervised machine learning techniques. It learns normal operational behaviour, detects previously unseen anomalies and predicts emerging failures based on factory topology. Instead of alerting operators to every symptom, Minerva groups related events into focused incidents with likely root causes. This reduces noise and helps teams distinguish between isolated glitches and systemic problems.

Affected versions

Minerva is a new release, not a patch for existing software flaws. It is available as part of NVIDIA’s ecosystem for AI factory operations. Several hardware vendors, including Dell Technologies, Lenovo and Supermicro, will offer the platform within their respective operations portfolios. The system integrates with external agentic AI tools and infrastructure application programming interfaces for automated remediation.

Mitigation

• Deploy Minerva to unify telemetry from GPUs, networks and storage into a single view.
• Configure the platform to learn normal behaviour patterns for your specific factory topology.
• Use the native AI assistant to investigate cluster health through natural language queries.
• Integrate with automation systems to implement recommended fixes manually or programmatically.

Why it matters

Large AI clusters generate immense data volumes that can overwhelm traditional monitoring. Teams usually struggle to correlate hardware faults with workload performance across disparate systems. This platform consolidates those signals, allowing operators to see the whole picture. Check whether your current tools can map dependencies between compute and network layers. Understanding these connections helps prevent small issues from cascading into major outages.

What to do and how to stay safe: NVIDIA

• Review your current telemetry collection to ensure it covers both infrastructure and workload metrics.
• Evaluate whether your monitoring tools can correlate events across different hardware layers.
• Plan for integration with automated remediation systems to reduce manual response times.
• Monitor for new observability features that can predict failures before they impact production.

Step-by-step guide: Stop Sensitive Data Leaks in AI Chatbots: Priority Controls

General security guidance from the Vector Bulletin newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

What is NVIDIA Minerva?

It is an AIOps platform that brings together telemetry from infrastructure and AI workloads to provide actionable insights for AI factories.

Which companies are using Minerva?

IREN is currently using it, while Aolani, Bitdeer AI, IBM, Jio, Lambda, Quake AI and Tesla plan to deploy the platform.

How does Minerva handle data?

It analyzes billions of data points per hour using machine learning to detect anomalies, predict failures and correlate related events.