Concepts

Platform Primitives

Instance Types

Spheron GPU offerings differ on two dimensions: interruptibility and hardware isolation.

Spot instances run as VMs and can be reclaimed by the provider at any time. Dedicated instances carry a 99.95% SLA and are not reclaimed once deployed.

Hardware isolation applies only to Dedicated instances, which come in three sub-types:

• VM: Isolated virtual machine on shared physical hardware. The default for most GPU offers across all providers.
• Bare Metal: Full physical server with no hypervisor. Identified by the BAREMETAL suffix in the GPU type name on the dashboard.
• Cluster: 8-GPU bare-metal server with a high-speed hardware interconnect. Supports InfiniBand (3.2 Tbps) or Ethernet (100 Gbps). Identified by the CLUSTER designation in the offers list. Currently available on Voltage Park only.

Cluster is the only sub-type that supports multi-GPU distributed training and the Kubernetes addon.

All instance types enforce a 20-minute minimum runtime before manual termination.

Regions & Providers

Spheron sources GPU capacity from six providers: Voltage Park, DataCrunch, TensorDock, Sesterce, Spheron AI, and Massed Compute. GPU availability, volume support, and startup scripts vary by provider.

Persistent volumes are supported on:

• Voltage Park (Dallas, TX): up to 64 TB, NFS-backed, one volume per instance
• DataCrunch (EU Finland: FIN-01, FIN-02, FIN-03): up to 10 TB, NVMe_Shared NFS-backed, can attach to multiple instances within the same datacenter region

The Kubernetes addon is exclusive to Voltage Park Cluster instances on H100 SXM5 offers.

Networking

Each instance gets a dedicated public IP for its deployment lifetime. There is no shared NAT or port forwarding. The IP address and SSH connection command are shown in the instance details panel in the dashboard.

All ports are open by default. Access unauthenticated services (Jupyter, TensorBoard) via SSH tunneling or VS Code Remote port forwarding. Do not expose unauthenticated services directly on the public IP.

Teams

Teams let multiple users share a credit pool, SSH keys, and volumes. Three roles are available: Owner, Admin, and Member. Only Owner and Admin can manage billing. For programmatic access with team scoping, see the API Reference.