Compute
Storage, networking and cleanup
Stopping a task, removing compute and retaining its outputs are distinct lifecycle decisions in SAVRN Cloud. This guide follows those relationships in the preview and explains the storage, networking and billing evidence a connected service would need before declaring cleanup complete.
SAVRN Cloud · Coming soon · Public preview · Browser-local simulation · No live compute or payments
Inspect what the demonstration retains
Local application records persist in browser localStorage. Outputs, fixture datasets and receipts shown in Datasets, Work Orders or Usage are part of that demonstration state. They are not backed by a production object store.
Use the available download actions when you want to retain a labeled sample for review. A download is a local file, not a backup service or a retention guarantee. Changing a display preference in Settings does not establish enforced deletion of external data.
Follow cleanup deliberately
Review a resource in Deployments or Sandboxes. Deployment Drain requests retains its allocation; Delete allocation, then Delete local allocation, requires exact-name confirmation and retains a deleted record. Sandbox Stop task concerns its sample task. Inspect the resulting state and activity. Resetting the entire demonstration is a separate action from stopping one resource; preserve any samples you need before reset.
Before connected services launch
Define ephemeral working storage separately from retained datasets, checkpoints and final artifacts. Document volume lifetime, region, encryption, access, retention, deletion and transfer charges. Persist promised artifacts before removing a worker. Block unintended access to host files and management networks. Destroying compute may leave storage or networking charges behind; verify all related resources and reconcile the final bill. A stopped application or powered-off instance is not sufficient evidence that billing ended.
Build toward the work that matters.
Tell SAVRN what your institution needs to run, who reviews the results and where its data must stay. That workload defines the next service to qualify.
Discuss an AI project