İçeriğe geç
Tres Teknoloji
Solutions / Disaster recovery

Up in the second region within minutes when disaster strikes

Disaster recovery as a service (DRaaS): we replicate your critical systems to a second region and, with low RPO/RTO targets, cut downtime from hours to minutes.

Requirements met
Second-region replication Continuous data replication between tr-ist-1 ↔ tr-izm-1.
RPO / RTO commitment Data loss and recovery time targets are defined in the contract.
Regular drills The recovery scenario is tested and reported once a quarter.
Data in Turkey The primary and secondary regions are within Turkey.
RPO 1-5 min Minimum data loss
RTO 15 min Maximum recovery time
2 regions Geographically separate replica
1 per quarter DR drill
Challenges

Taking a backup is one thing; being able to recover from a disaster is another

Most organizations have backups, but how long recovery from that backup takes and how much data is lost has not been tested. Disaster recovery is not a file copy but a rehearsed recovery process whose duration has been measured.

Continuous replication

Critical servers and databases are continuously copied to the second region.

Automatic failover

If the primary region is lost, the workload is brought up in the second region.

Measured RPO/RTO

Data loss and recovery time targets are defined and measured in drills.

Failback

Once the primary region recovers, traffic is moved back in a controlled manner.

Recommended architecture

A typical DRaaS architecture that replicates the production load in the primary region to a second region; whether the application is on the Tres cloud or in your own environment.

Primary region
Production servers Managed database Object storage
Replication
Block replication Database streaming Snapshot
Second region
Standby servers Replica database Load balancer
Management
Failover automation DR drill report Logchase Monitoring
Customer reviews

What teams using disaster recovery say

“During a regional failure our system came up in the second region in 12 minutes; by the time we found out, customers were still placing orders.”
TE Tolga Ersoy Infrastructure Director · An e-commerce platform
“Thanks to the quarterly drill report, we can prove to management that recovery actually works.”
SK Sibel Kaya Business Continuity Manager · An insurance company
“Seeing RPO and RTO clearly in the contract turned out to be the most useful thing in audits.”
UD Uğur Demir Information Security Manager · A financial institution
S.S.S.

Frequently asked questions about disaster recovery (DRaaS)

RPO refers to the maximum acceptable data loss (e.g. 15 minutes), while RTO refers to the maximum acceptable recovery time (e.g. under one hour). These targets are set according to your criticality, defined in the contract and measured in drills.

Backup copies data; disaster recovery is a rehearsed process that guarantees how long and with how much loss you will recover from that data. DRaaS automates this recovery with standby systems in a second region.

The replica is kept in a second region within Turkey (e.g. tr-izm-1). Both the primary and secondary regions are within Turkey; your data does not leave the country.

Both are possible. Automatic takeover is defined for critical scenarios; when you want a controlled switchover, it is triggered manually from the panel or under engineer supervision.

The recovery scenario is tested with a drill once a quarter; the measured RPO/RTO values are reported. This way you can be sure the process works before a disaster actually happens.

Yes. Whether the source is on the Tres cloud, in your own data center or with another provider; replication is set up over a site-to-site VPN and the second region is kept ready on the Tres side.

Once the primary region returns to normal, the changes accumulated in the second region are synced back and traffic is moved to the primary region with a controlled failback; this switchover is also planned with low downtime.

A case from this sector An e-commerce platform got through a regional failure in 12 minutes

During an infrastructure failure in the primary region, the workload was automatically failed over to the second region. Customers only saw a brief slowdown; the order flow returned to normal in 12 minutes.

12 min Actual recovery time
< 5 min Measured data loss
0 Lost orders

Talk to a team that knows your sector

A solution architect who has run projects in that sector joins the meeting. In the first meeting we produce an architecture draft and a cost range.

Schedule a meeting