DATA PROTECTION AND CONTINUITY

Backup and Disaster Recovery

A backup job that looks successful does not on its own prove that the application can be restored. We define acceptable data loss and downtime for critical systems and test backups against those targets. The written disaster recovery plan and drill records are prepared in the Compliance team’s DRP service.

SERVICE FOCUS
We test that backups can actually be restored

VeeamRPO / RTO3-2-1-1-0Recovery testing

Concrete goals · Documented work · Controlled implementation

WHEN IS IT NEEDED?

If backups are taken but a restore has never been tried

If any of the following applies, your backup design should be revisited against RPO and RTO targets.

01 / SITUATION

Single-copy risk

When all backups are on the same system or reachable with the same credentials.

02 / SITUATION

Untested recovery

When restores are not tested regularly and who does what is not written down.

03 / SITUATION

Unclear retention

When retention periods and critical data priorities don’t match business requirements.

SCOPE OF SERVICE

The four areas we cover in data protection

We don’t set the scope by software. Each system’s acceptable data loss (RPO), acceptable downtime (RTO) and ransomware risk define it.

01

Protection scope

Classifying VMs, files, databases, configurations and cloud workloads.

02

Backup architecture

Independent copies, offline/immutable options, network separation and access control.

03

Recovery planning

Service dependency order, recovery targets and alternative working scenarios.

04

Verification

Controlled restore tests at file, VM and application level.

ENGINEERING APPROACH

A backup success rate is not a recovery guarantee

RPO and RTO targets come from the business impact of the application. Files, virtual machines, databases and cloud services each need different recovery methods.

DECISION POINT

We first set RPO and RTO targets for each system, then choose the backup architecture and products to meet them. The work is completed by testing restores on selected systems.

Let’s review your infrastructure together ↗

RPO AND RTO

Not every system needs the same protection

An hour of data loss in an accounting database is a serious problem, while a day’s loss in an archive folder is often acceptable. Setting targets per system lets you get the cost and the backup window right.

SystemExample RPOExample RTOMethod
ERP / accounting database15 min – 1 hourA few hoursFrequent backups, transaction log backups; replication if needed
File server24 hours1 business dayNightly backup. Intraday shadow copies (VSS) are only a quick file restore option; they don’t count as an independent backup.
Domain controller24 hoursA few hoursNightly system state backup. A second domain controller keeps the service available; it does not replace a restore backup or a forest recovery procedure.
Microsoft 365 mail and files24 hoursDepends on the item type, the chosen recovery method and test results.A separate Microsoft 365 backup. Retention policies define how long data is kept; they don’t replace a backup.

The values in the table are examples. Real targets are set with business units by discussing what it costs the business when the system stops. Example times are not a service commitment. RTO covers bringing the application up, verifying dependent services and users being able to work again.

WORKING MODEL

A sequential process from backup to recovery plan

The work isn’t closed until recovery time has been measured.

01 / STAGE

Criticality

We classify systems by business impact and tolerance for data loss.

02 / STAGE

Design

We define copy placement, retention and security policies.

03 / STAGE

Implementation

We configure backup jobs suited to each workload.

04 / STAGE

Test

We verify recovery time and data integrity in selected scenarios.

DELIVERABLES

What is in place at the end of the data protection work?

A backup policy, retention periods, an immutable copy design and restore test records. Who does what, and in which order, during an outage is in writing.

RPO/RTO and criticality matrixBackup scope and retention policyRecovery order and runbookTest record and improvement findings

FAQ

Frequently asked questions about backup and disaster recovery

01

What is the 3-2-1-1-0 approach?

Three copies of the data are kept, on two different types of media, with one copy at a separate location. At least one copy must be offline, air-gapped or immutable. The final zero means no errors in backup and restore verification.

02

Does a backup system replace a disaster recovery plan?

No. Network, identity, application dependencies, responsible people and tested recovery procedures are also needed.

03

Do copies stored on a NAS count as backups?

A NAS can be used as a backup repository. However, a single, permanently connected NAS copy may be exposed to the same access rights or the same physical risk as production systems. We separate access and plan an independent copy.

04

Is a separate backup needed for Microsoft 365?

Microsoft is responsible for keeping the infrastructure running; long-term recoverability of the data largely depends on the customer’s configuration. If the restore window you need is longer than retention policies can cover, we recommend a separate Microsoft 365 backup.

05

Should backups be kept on site or in the cloud?

They complement each other rather than being alternatives. A local copy is used for fast restores, and the cloud or a second location for off-site protection. How long it takes to download data from cloud storage over your internet line must also be part of the RTO calculation.

LET’S PLAN TOGETHER

Let’s test together whether your backups really restore

Let’s talk about your critical systems, acceptable downtime and current backup setup, and then write a recovery plan that can be tested.

We set up backups and run backup checks remotely across Türkiye; when a backup appliance installation or an on-site restore test is needed, we schedule the work at the customer’s premises. Our service areas.