Bitdeer AI Brings NVIDIA Minerva's AI-Native AIOps For Maximizing AI Factory Yield
Bitdeer AI is pleased to announce that we now support NVIDIA Minerva, NVIDIA's new AI-native, end-to-end AIOps software platform purpose-built to maximize the availability and performance of AI Factory infrastructure. By adding NVIDIA Minerva to our AI Factory operations, Bitdeer AI is extending our commitment to delivering highly reliable, high-yield GPU infrastructure for AI training and inference at scale.
What is NVIDIA Minerva?
NVIDIA Minerva harnesses metric, log, and topology telemetry from both infrastructure and AI workloads, applying a broad range of purpose-built AI/ML algorithms and agentic techniques to turn a flood of raw telemetry into proactive, automated action - maximizing AI Factory yield, including GPU hours, goodput, cluster uptime, and token production.
It can be deployed as a standalone platform and is integrated with NVIDIA DSX OS.
A full-stack view no point tool can match
NVIDIA Minerva correlates telemetry across NVIDIA GPUs, NVIDIA ConnectX SuperNICs, NVIDIA BlueField DPUs, NVIDIA NVLink scale-up switches, NVIDIA Spectrum-X East-West and North-South Ethernet, NCCL, schedulers, and AI workloads. It transforms this telemetry into a dynamic knowledge graph that maps the cluster and the jobs running on it. That full-stack context lets NVIDIA Minerva connect a single anomalous interface, among hundreds of thousands, to the exact training job it affects.
NVIDIA Minerva analyzes billions of data points per hour in real time, learns normal behavior, predicts anomalies before they become incidents, and auto-correlates events across layers - so operators can focus faster on fewer, highly relevant issues. An MCP server and Sherlock, Minerva's platform-native multi-agent AI assistant, let operators triage and debug through natural-language conversations grounded in live telemetry.
Proven results
In NVIDIA's own internal AI Factory testing , NVIDIA Minerva has shown meaningful gains: an increase in Mean Time Between Interrupts through predictive maintenance, a decrease in Mean Time to Repair (MTTR) through automated root-cause analysis, an increase in actionable insights through advanced noise reduction, and an increase in effective GPU hours that directly boosts token production. In one case, NVIDIA Minerva reduced MTTR for an NVMe power fault from a multi-day outage to six hours; in another, it collapsed 81 noisy events into a single alert, saving 54 hours of cluster uptime.
Why it matters for Bitdeer AI
As Bitdeer AI scales its AI Factory infrastructure to serve AI training and inference workloads, every GPU hour counts. NVIDIA Minerva's full-stack observability, predictive maintenance, and agentic troubleshooting help us catch degradations before they impact jobs, cut repair time, and keep more GPUs productive for our customers, preserving data sovereignty, and built to scale from small clusters to AI Factories with hundreds of thousands of GPUs.
Bitdeer AI's support for NVIDIA Minerva reflects our ongoing commitment to building enterprise-grade AI infrastructure on the latest NVIDIA technologies.