Adding GPU Servers to an Existing OpenMetal Deployment

Q: Can I add GPU servers to my existing OpenMetal cloud or bare metal deployment?

Yes, you can add NVIDIA RTX Pro 6000 or H200 GPU servers to an existing OpenMetal Hosted Private Cloud or bare metal deployment, on the same private network as your current compute and storage.

Explore GPU servers

You do not need a greenfield project. GPU nodes are provisioned as dedicated single-tenant bare metal servers and joined to your existing private VLANs, so they sit next to your application servers, databases, and Ceph storage with low-latency, unmetered east-west traffic, under one provider and one billing model.

Choose the GPU to match the workload: attach RP6000 nodes for cost-efficient training and high-throughput inference, or H200 nodes for the largest-memory, bandwidth-bound work, and mix them as needed. Common patterns include adding local inference next to an application cluster (no public round-trip), adding training capacity to a bare metal footprint, and placing GPU nodes next to vector stores and Ceph object storage for RAG.

Added GPU capacity carries the same fixed monthly pricing and included egress as the rest of your deployment.