The classic ticket: “we need one file from last Tuesday’s backup of VM X” — and the service is down. Full-VM backup is still the #1 disaster-recovery use case. But when you need just one file, sub-par solutions make you mount entire VM disks with unsupported open-source tooling (libvirt, guestfs) — which doesn’t even read Windows disks by default. Minutes to hours of trial and error, mid-outage. No Gartner leader works that way.
NGBackup uses the state of the art: the backup indexes itself as the job finishes. Numbers measured in production (~80 vSphere VMs, Jul 2026):
- ⚡ 17 GB VM image opened in ~1.3 s (without the map: 20 min — ~900× slower)
- ⚡ One file out in 0.6 s, SHA-256 verified
- ⚡ NTFS, ext and btrfs parsed in userspace — no disk mounting, no kernel exposure
- ⚡ Chain-aware: browse any incremental point (Full + CBT deltas)
- ⚡ Agentless, read-only, ACL-gated and audited
Works across your whole fleet: the NGBackup VM plugins — VMware vSphere, Proxmox VE, Microsoft Hyper-V, Nutanix AHV, Apache CloudStack and OCI Compute — all write the same self-indexing image format.
And full-VM DR stays first-class: Granular Restore is one of the six recovery modes in Restore Studio — same backup, one click.
👉 Full article: Single-item restore is not Russian roulette
👉 Try NGBackup for free: download the free version (DEB/RPM repositories — 1-minute signup) or talk to our team and book a free POC.
Disponível em: 
Português (Portuguese (Brazil))
English
Español (Spanish)
