DRaaS (disaster recovery as a service) is a service in which a provider continuously replicates your servers to its cloud and, if your infrastructure fails, brings up an identical environment there so your company keeps operating while the original site is repaired. Unlike a backup, which recovers files, DRaaS recovers the entire operation.
This guide explains how it works, what it includes, how it's billed and when it makes sense to hire it instead of building your own secondary site.
What DRaaS is and how it differs from backup
It works like a hospital's backup generator: nobody uses it on a normal day, but if the power goes out it starts on its own and the operating rooms keep working. DRaaS does the same for your systems: while everything runs, it only copies each change; when something fails, it takes over.
| Backup | DRaaS | |
|---|---|---|
| What it recovers | Files, databases or entire devices | The entire operation: servers and applications running |
| Where you keep working | At your site, once you finish restoring | In an identical environment in the cloud, while your site is repaired |
| How long it takes to get back | Restoring can take hours or days | The environment comes up within minutes |
They don’t compete: backup protects your information and DRaaS keeps your operation running.
How it works: replication, failover and failback
A DRaaS service works in four stages:
- Replication. An agent installed in your infrastructure copies every change on your servers to the cloud, in near real time and around the clock. By the time something happens, your data is already somewhere else.
- Detection. A hardware failure, an attack or a human error doesn’t trigger anything on its own: the provider confirms the event with your team before starting recovery.
- Failover. It’s the switch from your site to the standby environment. The provider brings up an identical environment in the cloud and your team keeps working there.
- Failback. It’s the return. When your infrastructure is ready again, the updated data moves back to your site and the cloud environment is shut down.
Many services add point-in-time recovery: the ability to bring up your environment at an earlier moment, not just the most recent one. If encryption or an error damaged your data, you choose a point before the problem.
What a DRaaS service includes and how it’s billed
When comparing proposals, check that the service includes at least:
- Continuous replication of every protected server.
- Recovery tests on a regular basis that don’t affect production, with evidence of the measured times.
- A runbook: the step-by-step procedure the specialists follow on the day of the failure.
- Point-in-time recovery and how many days back it covers.
- Monitoring and support from the provider throughout the process.
Billing usually combines three parts: a fixed fee per server based on its size, storage for the replica, and the cloud hours you use only when you trigger a failover or an extra test. That last point matters: the rest of the time, your cost stays the same.
DRaaS or your own secondary site
The classic alternative is building a second data center. Here’s how they compare:
| Your own secondary site | DRaaS | |
|---|---|---|
| Investment | Duplicate servers, space, power and licenses | A monthly fee; cloud hours are billed when you activate |
| Time to get it ready | Months of purchasing and installation | A second site in the cloud, with nothing to build or buy |
| Who runs it | Your team, on top of daily operations | The provider’s specialists, with a runbook |
| Testing | Depends on your team having time to do it | Included in the service |
When it makes sense to hire DRaaS
DRaaS makes the most sense in these cases:
- Your ERP and billing can’t stop: if your day is measured in orders and invoices, your recovery should be measured in minutes.
- You want to get out of ransomware without paying: the environment goes back to a point before the attack and operations continue.
- Your audit or insurer requires a tested plan: frameworks like ISO 27001 and cyber insurance policies ask for documented tests.
- Your whole operation lives in one place: if that location goes down, your business stops completely.
Before hiring, define how long each system can be down and how much data you can afford to lose; these are your RTO and RPO targets.
If you don’t have a plan yet, start with how to prepare a disaster recovery plan.
What to ask before hiring DRaaS
- How many recovery tests are included, and what evidence do you get? Tests are the only way to know the plan works.
- How many days does point-in-time recovery cover? It defines how far back you can go if the problem took time to notice.
- Who declares the event, and how? It must be clear who decides to start recovery.
- What does the return to your site look like? A good service includes failback, not just failover.
- What are you charged when you activate recovery? Ask for the cost per cloud hour in writing.
With TecnetProtect DR, your servers are replicated continuously and, if they fail, your environment comes up in the cloud within minutes. It includes two recovery tests a year of up to 4 hours each, point-in-time recovery for the last 7 days and a runbook carried out by TecnetOne specialists. Cloud hours are only billed when you use them.
Frequently asked questions about DRaaS
It stands for disaster recovery as a service. A provider replicates your servers in its cloud and brings them up there if your infrastructure fails.
It’s the automatic or controlled switch from your main systems to a standby environment when something fails. The reverse move, back to your site once it’s ready, is called failback.
No. Backup keeps copies of your information; DRaaS keeps your systems running while you recover it. The recommendation is to have both.
Yes. You can go back to a point before the encryption and work from the cloud while your site is cleaned up.
Within minutes. The exact time depends on the number of servers and their capacity, and it’s defined in the service runbook; you confirm it in the recovery tests.
