Many teams deploy infrastructure without backup because they believe redundancy already covers them. RAID protects against a failed disk. High availability keeps a service running when a node goes down. Neither brings back data that has been deleted, corrupted or encrypted.
That gap matters more than most teams expect. In Leaseweb’s 2026 survey of Acronis users, 40% had already needed to restore from backup. For them, a restore wasn’t an edge case. It was a normal part of running infrastructure.
TL;DR
Redundancy keeps your systems running; backup brings your data back. They solve different problems, and you need both. Keep three copies of your data on two types of storage, with one off-site and protected against changes. Then test a restore every quarter, so you know it works before you need it.
Redundancy keeps you running. Backup brings you back.
Hardware redundancy makes a server fault-tolerant. RAID, for example, spreads data across several disks so one failure doesn’t take the server down. Service redundancy, or high availability, keeps a service online when hardware fails by moving it to another node.
Both are about uptime. And both copy whatever happens to your data, including the things you don’t want. Delete a table, and the deletion is mirrored instantly. Ransomware encrypts files, and every mirrored disk holds the encrypted version.
A backup works differently. It’s a point-in-time copy kept separate from production, so you can return to the state your data was in before something went wrong.
Five everyday failures redundancy can’t fix
- Human error. Every command you run and every configuration you deploy can delete or overwrite files. It happens to experienced teams too.
- A failed update or corrupted data. An update breaks the operating system, or a database becomes corrupted. RAID faithfully mirrors the damage.
- Ransomware. Attackers encrypt production data and anything else they can reach. Only a separate copy they can’t change gets you back without paying.
- Total hardware or site failure. Several disks or a controller fail at once, or an incident affects the whole facility. Redundancy inside one server or one site can’t help.
- An audit. Frameworks such as ISO 27001 and NIS2 expect you to show that you can recover data, not just that your systems stay online.
What a solid backup strategy looks like
The industry baseline is the 3-2-1 rule:
- 3 copies of your data: production plus two backups
- 2 types of storage, so one failure mode can’t take out every copy
- 1 copy off-site, outside your own infrastructure
Two additions make it hold up against today’s attacks. Keep the off-site copy immutable, so nobody can change or delete it for a set period, not even an administrator. And keep backups on a separate network with separate credentials, so a compromised domain account can’t reach them.
Finally, test your restores. A backup you’ve never restored is a backup you hope works. Restoring one critical workload into a test environment each quarter tells you whether it does, and how long it takes.
Where backups still fall short
Having a backup is the first step. Many teams that already have one still run into these gaps:
- Backups that are hours old. With a daily backup at midnight, a database crash at 11 PM costs 23 hours of transactions. Continuous backup of critical applications shrinks that loss to seconds.
- Ransomware that antivirus misses. Signature-based tools can’t recognize new ransomware strains, which can encrypt backups along with production data. Behavioral detection watches for unusual encryption activity and isolates backups before they’re hit.
- Backups deleted before the attack. Attackers increasingly go after backups first, and one compromised admin account can delete every recovery point. Immutable copies can’t be changed or deleted for a set period, not even by an administrator.
- Full restores for small problems. When only a few database tables are damaged, restoring a whole system or cluster means hours of downtime. Granular restores bring back just the files, emails or records you need.
- IT as the bottleneck. If every restore needs an engineer, recoveries queue up when several systems are hit at once. Self-service recovery lets authorized users restore their own workloads.
Since 1997, Leaseweb has run infrastructure for customers who want to decide where their data lives and who can change it. Backup is part of that control: a copy you can return to, stored in a location you choose.
If you run servers with Leaseweb, Acronis Backup is available for Dedicated Servers, VPS and Public Cloud. Plans now include the capabilities above, among them continuous data protection, immutable storage and granular recovery, at no extra cost, with backups stored in the Acronis data center you select.
Read next:
- What’s included in Acronis Advanced Backup
- The five steps ransomware takes to destroy your backups, and what stops each one
- Backup recovery testing
- Migration backup gaps
About the survey: Leaseweb surveyed customers using Acronis Cyber Protect on Leaseweb in June 2026. Between 40 and 43 customers answered each question. With a sample this size, the results describe this group of Leaseweb customers and aren’t statistically representative of all backup users.