Storage

On a self-hosted Arcsecond.local install, your data lives on your own infrastructure. There is no cloud bucket, no per-gigabyte billing and no storage quota — the only limit is the disk you give the install.

Where data is stored

When you run arcsecond setup, you choose a host directory as the SHARED_DATA_PATH. Everything Arcsecond.local persists lives under it:

• uploaded datasets and data files,
• generated 2D-image and 1D-spectrum previews,
• the astrometry cache and temporary working files,
• the hourly database backups (see Backups).

Because it is a plain directory on your machine, you back it up, snapshot it and size it with your own operational practice. The Backups and disaster-recovery guides cover what to copy off-host.

How data is organised

Data files are grouped into Datasets — virtual folders that collect files sharing a common source or interest, so they can be browsed and downloaded together. FITS and XISF files are parsed on ingestion: their headers are extracted for searching, and previews are generated.

Bringing in data you already have

You rarely start from a blank slate. Arcsecond.local can index data that already lives elsewhere — on an S3 bucket, a Dropbox, an FTP/SFTP server, another local disk, or an observatory archive — by attaching it as an External Storage. Attached storages are read-only: Arcsecond.local scans and parses them exactly as it does its own data, never modifying anything.