restic: Open-Source Backup Ledger 2026
restic is the most widely adopted tool in this Ledger: encrypted, deduplicated backups to almost any storage. A tool, not a platform: no central console.
Where restic put its effort, 2017 to 2026
Each square is one capability area in one year. The brighter it is, the more the area gained in the Open-Source Backup Ledger that year; 1, 2 and 3 mark that year's biggest pushes.
| ’17 | ’18 | ’19 | ’20 | ’21 | ’22 | ’23 | ’24 | ’25 | ’26 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Virtual and physical servers | ||||||||||
| Databases and applications | ||||||||||
| Recovery | ||||||||||
| Cyber resilience | 2 | 2 | ||||||||
| Storage efficiency and targets | 1 | |||||||||
| Long-term retention and tape | ||||||||||
| Replication and DR | 1 | |||||||||
| Security and compliance | ||||||||||
| New-platform support (VMware exit) | ||||||||||
| Management at scale and automation | 1 |
Biggest pushes overall: Replication and DR (+3), Cyber resilience (+1), Storage efficiency and targets (+1). Already strong in 2016: Storage efficiency and targets, Recovery, Long-term retention and tape.
Top focus each year: 2019 Management at scale and automation; 2020 Replication and DR; 2022 Storage efficiency and targets. 7 of 10 years show no gain in any area.
Part of the Open-Source Backup Ledger (edition 2026-10-02). How the scores work: methodology.
Snapshot
- Kind: Backup tool
- Licence: BSD-2-Clause
- Backer: Community; sponsors include Backblaze, rsync.net and Wasabi
- Paid edition or support: No vendor support
- Latest release: 0.19.1, 2026-07-05
- Next signal: Memory reduction in prune, faster uploads, more verification
- Archetype: Community tool, broad base
- Evidence grade: A. Release dates verified against GitHub; health figures re-read from the API on 2026-10-02.
Scope of the scores: A backup tool: single binary, no central server or scheduler (wrappers such as Backrest add them).
Scorecard
| Dimension | Score | Direction |
|---|---|---|
| Capability today | 56.4 / 100 | higher is better |
| Momentum since end of 2022 | +0.0 points | higher is better |
| Credibility | 21.0 / 25 | higher is better |
| Sustainability risk | 9.0 / 25 | lower is better |
| Technical lock-in | 4.5 / 25 | lower is better |
| Support and ecosystem | 13.0 / 20 | higher is better |
| AI leverage | 1.0 / 20 | higher is better |
| Ledger Index | 58.1 / 100 | balanced weights |
Capability by domain
Scores 0 to 5 against fixed anchors; total is weighted to 100.
| Domain (weight) | 2016 | 2019 | 2022 | Today |
|---|---|---|---|---|
| Virtual and physical servers (12) | 2.5 | 2.5 | 2.5 | 2.5 |
| Databases and applications (12) | 1.5 | 1.5 | 1.5 | 1.5 |
| Recovery (14) | 3 | 3 | 3 | 3 |
| Cyber resilience (14) | 2.5 | 3 | 3.5 | 3.5 |
| Storage efficiency and targets (10) | 3.5 | 3.5 | 4.5 | 4.5 |
| Long-term retention and tape (6) | 3 | 3 | 3 | 3 |
| Replication and DR (10) | n/a | n/a | 3 | 3 |
| Security and compliance (8) | 3 | 3 | 3 | 3 |
| New-platform support (VMware exit) (8) | 1 | 1 | 1 | 1 |
| Management at scale and automation (6) | 2 | 3 | 3 | 3 |
| Total / 100 | 44.4 | 47.0 | 56.4 | 56.4 |
Year by year: 2016: 44.4 · 2017: 44.4 · 2018: 44.4 · 2019: 47.0 · 2020: 53.0 · 2021: 53.0 · 2022: 56.4 · 2023: 56.4 · 2024: 56.4 · 2025: 56.4 · 2026: 56.4
What moved the score
| Year | Domain | Change | Points | Trigger |
|---|---|---|---|---|
| 2019 | Cyber resilience | 2.5 → 3 | +1.4 | Append-only REST server |
| 2019 | Management at scale and automation | 2 → 3 | +1.2 | JSON output and REST server |
| 2022 | Cyber resilience | 3 → 3.5 | +1.4 | Works with S3 Object Lock; integrity checks |
| 2022 | Storage efficiency and targets | 3.5 → 4.5 | +2.0 | 0.14: compression; Azure, GCS, rclone back ends |
Direction, 2023 to 2026
Each item carries one theme and one driver (V vision, C customer need, M market window, U upstream, P portfolio).
| Date | Item | Theme | Driver |
|---|---|---|---|
| 2023-07 | 0.16: dry-run prune, repository repair | D3 Recovery | C |
| 2024-07 | 0.17: faster restores | D3 Recovery | C |
| 2025-03 | 0.18: S3 Object Lock verification | D4 Cyber resilience | M |
| 2026-07 | 0.19: multi-threaded restore memory optimisation | D3 Recovery | C |
- Centre of gravity: Recovery (3), Cyber resilience (1)
- Driver mix: Vision 0%, Customer need 75%, Market window 25%, Upstream 0%, Portfolio 0%
Credibility: 21.0 / 25
| Criterion | Score | Basis |
|---|---|---|
| Cadence | 4 | Two to five releases a year |
| Roadmap transparency | 4.5 | Public milestones and issues |
| Say-do | 4.5 | 7 of 7 items shipped, 4 to 14 months |
| Velocity | 4 | Steady |
| Lifecycle stability | 4 | Format v2 introduced without forcing upgrades |
Say-do record, 2023 to 2026: 7 items since 2023, all shipped within 4 to 14 months.
Sustainability risk: 9.0 / 25 (lower is better)
| Criterion | Score | Basis |
|---|---|---|
| Maintainer concentration | 3 | Co-lead maintainer writes 57 to 76% of commits; two leads |
| Backer health | 3 | No company; storage-provider sponsors |
| Licence-change risk | 1 | Permissive, distributed copyright |
| Abandonment signs | 0.5 | Active |
| Security response | 1.5 | Active fixes |
Project health: 36,373 stars, 1,901 forks; 475 open issues (median 46 days), 133 open PRs (11 days). 11 to 26 distinct authors a year.
Technical lock-in: 4.5 / 25 (lower is better)
| Criterion | Score | Basis |
|---|---|---|
| Formats | 2 | Own repository format, documented |
| Export path | 1 | Restore, dump or mount |
| Stack coupling | 0.5 | Single Go binary |
| Hardware / cloud coupling | 0 | Any local or cloud target |
| Skills and tooling coupling | 1 | Common skills |
Support and ecosystem: 13.0 / 20
| Criterion | Score | Basis |
|---|---|---|
| Paid support | 1.5 | No vendor support |
| Consultants and partners | 2.5 | Used widely by consultancies |
| Documentation | 4 | Good documentation |
| Community and packaging | 5 | Largest community here; in all distributions; Backrest and resticprofile around it |
AI leverage: 1 / 20
| Strand | Today | Latest step |
|---|---|---|
| AI for protection / protecting AI | 0 | none |
| AI platform stack | 0 | none |
| AI-assisted operations | 0 | none |
| Agent openness | 1 | 2019: JSON output |
No AI features.
Reading
restic is the most widely adopted backup tool in this Ledger: encrypted, deduplicated, compressed backups to almost any storage, delivered reliably by a responsive project. It is a tool, not a platform: no central console, database agents or VM integration of its own.
Open questions and evidence caveats
- Still a 0.x version number after ten years; the repository format is stable in practice.
Related reading
More from Open-Source Backup Ledger 2026
- Overview: all platforms side by side
- Open-Source Backup Ledger Methodology: How Every Score Is Built
- Amanda Community: Open-Source Backup Ledger 2026
- BackupPC: Open-Source Backup Ledger 2026
- Bacula Community: Open-Source Backup Ledger 2026
- Bareos: Open-Source Backup Ledger 2026
- Barman: Open-Source Backup Ledger 2026
- BorgBackup: Open-Source Backup Ledger 2026
- Burp: Open-Source Backup Ledger 2026
- Kopia: Open-Source Backup Ledger 2026
- Percona XtraBackup: Open-Source Backup Ledger 2026
- pgBackRest: Open-Source Backup Ledger 2026
- Proxmox Backup Server: Open-Source Backup Ledger 2026
- Relax-and-Recover: Open-Source Backup Ledger 2026
- UrBackup: Open-Source Backup Ledger 2026
- WAL-G: Open-Source Backup Ledger 2026
- Xen Orchestra: Open-Source Backup Ledger 2026