Platform

Cloud Storage

A drive in the cloud for your files — the same job Google Drive, Dropbox, or Box does, on the storage platform that boots operating systems. Upload anything, browse and share it from the web, mount it as a folder on your desktop, and when you need a real computer around your files, open the drive inside a brintOS virtual machine.

The third kind of repo

brintOS storage comes in three classes, all living on the same account, the same permissions model, and the same storage substrate:

  • Machine images — bootable root filesystems with a Launch button.
  • Git repos — bare repositories you clone and push with stock git.
  • Cloud storage drives — plain folders of files. Documents, photos, datasets, build artifacts, backups: anything you'd put in a sync folder.

A drive has no Launch button, because there's nothing to boot — it holds your files, not an ext4 disk image. That's by design: not everything you store is an operating system. And when you do want a running system around your files, you don't convert anything — you open the drive from inside a virtual machine and work on it with real Linux tools.

Use it like a drive

  • Web first. Drag files or whole folders into the browser to upload. Browse, preview, search, and download from any device with a URL bar — images render, Markdown renders, text gets syntax highlighting.
  • Mount it on your desktop. The brintos CLI mounts any drive as a folder on your Mac, Linux, or Windows machine. It behaves like a local directory; reads stream on demand and writes sync up in the background. No "sync client" babysitting a copy of everything.
  • Open it in a VM. Any drive can be attached to a running brintOS machine. Your photo archive meets imagemagick; your dataset meets pandas; your backup meets rsync — without downloading a byte to your laptop.
  • Share like you expect. Drives are public, private, or internal to your organization — the same visibility model as every other brintOS repo. Share a whole drive or a link to a single file or folder.

Versioned like a repo, not like a backup

Because drives sit on the same content-addressed storage as machine images, every change is a snapshot, not an overwrite. The file you replaced last Tuesday still exists; restoring it is a checkout, not a support ticket. There's no recycle-bin retention window to race against — history is part of the drive.

The substrate underneath

  • Content-addressed and deduplicated. Every file is hashed with SHA-256 and stored once. Ten copies of the same video across your drives — or across your whole team — cost one copy of the bytes.
  • Transparently compressed. Compressible files are stored compressed and served decompressed. You see your files; the quota sees fewer bytes.
  • Edge cached, globally. Public content is fronted by a global CDN, so a shared file downloads from a server near the person you shared it with, not near us.
  • Streamed, not copied. Mounts — desktop and in-VM alike — fetch only the blocks that are actually read. Opening a 200 GB drive takes milliseconds; you pay the I/O for the files you touch.

One quota, no surprises

Drives draw on the same cloud storage allowance as your machines and git repos — 10 GB free, up to 1 TB and beyond on paid plans, with add-ons past that. There's no per-file size gotcha and no per-drive count limit. Public drives are free to share: people downloading your public files don't burn anyone's bandwidth allowance.

Bring-your-own-bucket

On Enterprise, the entire blob store can live in a bucket inside your AWS account. brintOS signs reads and writes through a scoped IAM role, but your bytes never leave your region or your billing line. Get in touch if you need single-tenant storage, customer-managed keys (CMK), or in-region residency guarantees.