Service Details

Cloud Disaster Recovery Planning

Know how long recovery takes and how much data you could lose, then test it. Backup, restore and failover for AWS and Azure, written down as a runbook.

A Backup You Have Never Restored Is a Hope

Many teams have backups switched on and no evidence they work. Others run production and test on one database in one availability zone. Disaster recovery planning starts with two numbers for each system: how quickly it must be back (RTO) and how much data you can afford to lose (RPO). From there we design the cheapest setup that meets them, restore from your backups to prove they work, and write the runbook your team follows on a bad day.

Recovery Time Objective is how long a system can be down. Recovery Point Objective is how much recent data you can afford to lose. A payments system and an internal reporting tool need very different numbers, and the numbers decide the cost of the design.
Only if you have restored from them. We run a test restore, time it, and check the data. Backup jobs that fail without anyone noticing, and restores that take longer than expected, are the things a test is there to catch.
Often not. For many platforms, Multi-AZ databases and tested backups copied to another region meet the recovery targets at a fraction of the cost of a full standby. We state what each option costs so you can choose.
Usually it adds some cost, and we say so openly. In one review we recommended Multi-AZ for a database shared by production and test, and stated the extra cost against the single point of failure it removes.
With a rehearsed plan. For Amazon RDS we use Blue/Green Deployments so the old environment stays available until the new one is validated, and users see a maintenance page during the short switchover.
Backup, recovery and availability controls are part of SOC 2 when availability is in scope, and auditors ask for evidence that restores are tested. The runbook and test records from this work serve as that evidence.

Recovery Options, Cheapest First

Chosen to match your recovery targets, not to maximise spend.

Backup and restore

Automated backups, copied off-account or cross-region, with restores tested on a schedule.

Multi-AZ

Databases and services that survive the loss of one availability zone.

Warm standby

A scaled-down copy in a second region, ready to take traffic.

Safe upgrades

Blue/Green database upgrades with rollback kept until sign-off.

How We Plan Recovery

Targets first, then design, then proof.

01

Targets

RTO and RPO agreed for each system.

02

Design

The lowest-cost setup that meets them.

03

Test

A real restore, timed and checked.

04

Runbook

Written steps for the day it is needed.

Recovery you have seen work

The output is a tested restore and a runbook, not a diagram of what should happen.

Tested

Restores are run and timed.

Costed

Every option comes with its monthly cost.

Written down

A runbook anyone on the team can follow.

Monitored

Alerts when a backup job fails.

Related case studies

Start with a free cloud audit

Read-only access, a written list of findings, and no obligation to continue.

Book a free cloud audit