Backup Policy

From HPCwiki
Revision as of 11:43, 18 June 2026 by Haars0011 (talk | contribs) (IA migration §6: split/synthesize Filesystems backup info into Backup Policy (via create-page on MediaWiki MCP Server))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Different storage areas on Anunna come with very different backup guarantees. Knowing which is which is essential: some areas are never backed up, and data lost there cannot be recovered. This page summarises the situation; the page for each storage area gives more context.

Storage Backed up? Notes
/home Yes — to tape, restorable from roughly a week of history. Good for configuration files and small results you can't afford to lose.
/lustre/nobackup No. The default Lustre area; cheaper, but lost data cannot be recovered. Keep your own copies of anything important.
/lustre/backup Yes — a second copy on a separate machine. A disaster-recovery copy, not a versioned backup. It may help revert a mistake if you report it very quickly, but there is no guarantee.
/lustre/scratch No. Temporary area; files may be purged automatically (typically after 30 days).
/archive Yes — to tape, restorable from roughly a week of history (same schedule as /home). Warm storage on the login nodes, for WUR users.

Recommendations

  • Treat /lustre/nobackup and /lustre/scratch as expendable. Anything you cannot regenerate should also live somewhere that is backed up, or be archived to tape — see Archival Storage.
  • A "backup" tier protects against hardware disaster, not against your own mistakes. If you delete something by accident, report it immediately; recovery is best-effort.

See also