Skip to main content

​
Private Cloud FAQ

Private Cloud provides dedicated, bare-metal GPU clusters reserved exclusively for your organization. Unlike Public Cloud (shared, on-demand), Private Cloud gives you single-tenant infrastructure with guaranteed capacity on a fixed monthly contract.
NVIDIA H100 (80 GB), H200 (141 GB), B200 (192 GB), and B300 (288 GB). All with NVLink intra-node and InfiniBand inter-node networking. See Available GPUs for full specs.
The minimum is typically 16 nodes (128 GPUs). For smaller needs, consider Public Cloud GPU Instances which support 1–8 GPUs per instance with no commitment.
Typically 1–2 weeks from signed agreement to live cluster, depending on GPU type and location.
12 or 24 months. Longer contracts may offer better per-GPU pricing.
Yes. You can add nodes to your existing cluster subject to availability. Contact your account manager to discuss expansion.
Anything. You have full root access to bare-metal servers. Common setups include Kubernetes, Slurm, Docker, and direct bare-metal access. Runcrate can assist with managed Kubernetes or Slurm if needed.
North America (US West, US East, Canada), Europe (Germany, Netherlands, UK), and Asia Pacific (South Korea, Japan, Singapore, India, Vietnam). Availability varies by GPU type. Contact us for current options.
Fixed monthly rate based on GPU type, cluster size, and contract duration. Per-GPU-hour pricing is locked for the entire contract. No surprise fees, no usage spikes.
Yes. Private Cloud customers receive an uptime SLA and dedicated support. Details are included in your service agreement.
Absolutely. Many customers start on Public Cloud for prototyping and development, then move to Private Cloud when they need reserved capacity at scale. Our team can help plan the transition.
Email support@runcrate.ai with your GPU requirements. We’ll respond within 24 hours with a proposal.