Business Continuity & Disaster Recovery

Recovery Time Objective (RTO) answers: how long can this system be down before the business is seriously harmed? Recovery Point Objective (RPO) answers: how much recent work can we afford to lose?

An eight-hour RTO and a 24-hour RPO is a nightly backup restored to spare hardware. A fifteen-minute RTO and a fifteen-minute RPO is a very different architecture with local virtualization and continuous replication. The design follows the number the business chooses — and the cost follows the design.

• Image-level server backup with local and cloud copies
• File-level backup and Microsoft 365 data protection
• Cloud replication and geographic redundancy
• Local virtualization for fast failover of critical servers
• Immutable backup storage to survive ransomware
• Automated backup verification and boot testing
• Documented recovery runbooks per system
• Scheduled recovery drills with written results

Depending on architecture, failure type, configuration and recovery method, some recovery scenarios can bring a critical server back into service in roughly five minutes by booting a recent image locally. That is a real capability with real prerequisites — it is not a universal guarantee, and we will tell you which of your systems qualify and which do not.

Restoring a large database over a slow link, recovering a physical machine with unusual hardware, or rebuilding an environment after a full site loss are longer operations. Planning starts by being clear about which is which.

Backup jobs report success for all kinds of reasons that have nothing to do with whether the data is recoverable. We verify by actually booting protected images and restoring sample data on a schedule, and we document the result.

When a recovery drill fails, that is a good day: it failed on a Tuesday morning instead of during an incident.