DRI IT Disaster Recovery and Resilience 2 — Questions and Answers

Question 1: Which cloud-based disaster recovery model uses cloud infrastructure to maintain a continuously updated replica of production systems, enabling rapid failover?

Cold cloud backup
Pilot light configuration
Disaster Recovery as a Service (DRaaS) with active replication (Correct answer)
Tape-to-cloud archival backup

Correct answer: Disaster Recovery as a Service (DRaaS) with active replication

DRaaS with active replication continuously replicates production data and system state to cloud infrastructure, enabling rapid failover with minimal data loss and recovery times comparable to on-premises hot sites at a fraction of the capital cost.

Disaster Recovery as a Service (DRaaS) represents a cloud-era evolution of traditional recovery site models. In the most capable DRaaS configurations, enterprise-grade replication software continuously mirrors production servers, databases, and applications to cloud infrastructure. During a declared disaster, cloud instances are spun up from the replicated state within minutes, providing RPOs measured in minutes and RTOs measured in hours. This compares favorably to traditional warm site arrangements that required hardware pre-provisioning and days of recovery time. DRaaS eliminates the capital expense of dedicated recovery hardware and the operational overhead of maintaining a separate DR facility, replacing them with ongoing subscription costs. Key considerations include replication bandwidth requirements, testing frequency (most DRaaS providers offer non-disruptive testing), data sovereignty and compliance, and vendor reliability. DRaaS has become a mainstream DR strategy for organizations of all sizes.

Question 2: What is 'database shadowing' in the context of IT disaster recovery?

Creating encrypted copies of sensitive databases for compliance purposes
Running concurrent transactions against both the primary and backup databases in real time to maintain synchronization (Correct answer)
Maintaining a compressed archive of database transaction logs offsite
Monitoring database performance to detect anomalies before they cause failures

Correct answer: Running concurrent transactions against both the primary and backup databases in real time to maintain synchronization

Database shadowing involves simultaneously writing transactions to both the primary database and one or more shadow copies in real time, maintaining continuously synchronized duplicates that can be activated immediately if the primary fails.

Database shadowing is a specific technique for achieving very low RPOs (near zero data loss) by maintaining shadow copies of the primary database that receive all transactions simultaneously. Unlike asynchronous replication with replication lag, shadowing (also called synchronous mirroring) confirms that a transaction is written to both primary and shadow before acknowledging success to the application. This ensures the shadow copy is always current to the last committed transaction. The shadow database can be promoted to primary status within minutes if the original fails, with no data loss. Shadowing is resource-intensive, requiring duplicate hardware and storage, and the synchronous write overhead can impact performance, particularly for high-transaction-rate systems. It is most appropriate for databases with zero or near-zero RPO requirements—financial transaction systems, healthcare records, real-time inventory systems—where even minutes of data loss would be unacceptable.

Question 3: Which IT resilience technique distributes application workloads across multiple geographic locations, allowing each location to handle the full production load if others become unavailable?

Cold standby with manual failover
Active-active geo-redundant architecture (Correct answer)
N+1 server clustering within a single data center
Scheduled batch backup to a secondary site

Correct answer: Active-active geo-redundant architecture

Active-active geo-redundant architecture runs production workloads simultaneously across multiple geographic locations, each capable of handling the full load—enabling automatic failover with zero (or near-zero) downtime when a location fails.

Active-active geo-redundant architecture is the gold standard for applications requiring the highest levels of availability. Unlike active-passive (primary handles all traffic, passive is a standby), active-active configurations route production traffic to multiple locations simultaneously—typically using global load balancers that detect failures and redirect traffic within seconds. Each location maintains its own data synchronization, ensuring sessions and transactions can continue seamlessly. This architecture eliminates both single-site failures and the recovery time associated with activating a passive standby. The complexity and cost are substantial: data consistency across geographically distributed writes is a significant engineering challenge, requiring careful consideration of database consistency models (ACID vs. eventual consistency), conflict resolution for simultaneous writes, and latency management. Active-active is appropriate for tier-one applications where any downtime has immediate, significant business impact.

Question 4: In IT disaster recovery planning, what is the 'Recovery Time Actual (RTA)' metric used for?

Setting the initial RTO target during BIA
Measuring the actual time taken to recover a system during a test or actual incident, for comparison against the RTO target (Correct answer)
Calculating the financial cost of recovery operations
Determining the staffing requirements for the recovery team

Correct answer: Measuring the actual time taken to recover a system during a test or actual incident, for comparison against the RTO target

Recovery Time Actual (RTA) measures the real time taken to recover a system during an exercise or actual incident, enabling comparison against the RTO target to determine whether the recovery solution is meeting its performance requirements.

The RTA metric closes the feedback loop between planning and performance. RTOs are established during BIA based on business impact analysis—this is the target that drives recovery strategy design and investment. RTA is collected during DR tests, partial activations, and actual incidents by tracking the precise time from incident declaration to functional restoration of the system. When RTA exceeds RTO, the recovery solution is not meeting its performance requirements and remediation is needed: faster recovery procedures, additional pre-staging of resources, better-trained recovery teams, or upgraded recovery infrastructure. When RTA consistently meets or beats RTO with margin to spare, the organization may be over-investing in recovery capabilities for that system. Regular tracking of RTA through testing provides the empirical data needed to validate recovery investments, demonstrate program effectiveness to leadership, and drive targeted improvements where gaps exist between planned and actual recovery performance.

Question 5: What is the purpose of conducting a 'tabletop test' specifically for IT disaster recovery?

To physically test the activation of the alternate data center
To walk recovery team members through a disaster scenario to validate procedures, roles, and decision-making without activating actual recovery systems (Correct answer)
To benchmark IT system recovery time against industry standards
To test the physical security of the primary data center

Correct answer: To walk recovery team members through a disaster scenario to validate procedures, roles, and decision-making without activating actual recovery systems

An IT DR tabletop test involves walking key recovery personnel through a simulated disaster scenario to validate that procedures are understood, roles are clear, and decision-making sequences are logical—without the risk and cost of actually activating recovery systems.

Tabletop tests occupy the first tier of the DR exercise progression and are particularly valuable for IT DR because they expose procedural gaps, knowledge deficiencies, and coordination failures before committing to more expensive and disruptive functional tests. A typical IT DR tabletop presents a scenario (ransomware attack encrypts primary data center systems at 2 AM on a weekday) and walks participants through their response: Who gets called first? What criteria trigger a disaster declaration? Who has authority to activate the recovery site? In what sequence are systems recovered? How is the business notified? How are vendors engaged? Tabletops often reveal that contact information is outdated, that recovery sequence assumptions are incorrect, that decision authorities are unclear, and that team members are unfamiliar with their specific recovery responsibilities—all issues better discovered in a conference room than during an actual incident.

Question 6: Which of the following BEST describes 'immutable backups' and their importance to IT resilience?

Backups stored on systems that cannot be accessed by production network users
Backup copies that cannot be modified, deleted, or encrypted for a defined retention period, protecting against ransomware and accidental deletion (Correct answer)
Backup copies maintained in geographically separate locations
Backups that are automatically tested and verified upon creation

Correct answer: Backup copies that cannot be modified, deleted, or encrypted for a defined retention period, protecting against ransomware and accidental deletion

Immutable backups are copies that cannot be altered, deleted, or encrypted by any user or process—including ransomware—for a defined retention period, providing a guaranteed clean recovery point even when production systems and conventional backups are compromised.

The rise of ransomware has fundamentally changed backup strategy requirements. Traditional backup architectures, even with offsite copies, proved vulnerable because ransomware operators discovered they could target backup systems before encrypting production data—eliminating the clean recovery point. Immutable storage addresses this by writing backup data in a WORM (Write Once, Read Many) format, either using object storage with object lock features, purpose-built immutable backup appliances, or air-gapped systems with physical disconnection. During the retention period, no credentials—not even system administrators—can modify or delete the data. This guarantees a clean recovery point exists even when all other systems are compromised. Best practices require: testing immutable backups regularly to verify restorability, storing immutable backups in a separate security domain from production, and maintaining at least one copy in a different geographic location with different access credentials. Immutable backups have become a baseline IT resilience requirement for virtually all organizations.

DRI IT Disaster Recovery and Resilience 2

Which cloud-based disaster recovery model uses cloud infrastructure to maintain a continuously updated replica of production systems, enabling rapid failover?