Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.brightnode.cloud/llms.txt

Use this file to discover all available pages before exploring further.

Network volumes are Brightnode’s persistent storage layer. They live independently of individual machines, so your datasets, checkpoints, and shared assets remain available even when a Bnode is stopped or deleted.

​
When to use network volumes

Use a network volume when you need:
  • Persistent data that should outlive a single Bnode.
  • Shared storage for datasets, model weights, or generated artifacts.
  • A portable workspace that you can attach to new compute as needed.
  • Direct file management through the S3-compatible API.

​
Pricing

Network volumes are billed monthly by storage size:
  • $0.07 per GB for the first 1 TB.
  • $0.05 per GB above 1 TB.
If your account balance cannot cover storage costs, the volume may be terminated and Brightnode cannot recover the lost data. Back up critical information regularly.

​
Create a network volume

  1. Open the Storage page in the Brightnode console.
  2. Select New Network Volume.
  3. Choose the datacenter where the volume should live.
  4. Give the volume a descriptive name.
  5. Choose the size in GB.
  6. Select Create Network Volume.
You can increase network volume size later, but you cannot decrease it.

​
Use a network volume with Bnodes

Attach the volume during Bnode creation so your machine starts with shared persistent storage available.
  1. Go to the Bnodes page.
  2. Start a new Bnode deployment.
  3. Select the network volume you want to attach.
  4. Finish the rest of the deployment flow.
When attached to a Bnode, the volume is typically available at /workspace.

​
Access a network volume without a machine

Brightnode also provides an S3-compatible API for network volumes. This is useful when you want to upload datasets, download artifacts, or automate file management without starting a Bnode first. Use the S3-compatible API when you want to:
  • Preload large datasets before attaching the volume to compute.
  • Sync files from CI or local development environments.
  • Manage files with standard S3 tools such as AWS CLI or Boto3.

​
Share data safely

Multiple Bnodes can read from the same volume, but concurrent writes to the same files can still cause conflicts. If your workflow writes to shared paths, add your own locking or job coordination.