712-50 Question 288
Single answerDesign Backup and disaster recovery strategies for cloud computingA global company is migrating a customer-facing analytics platform to a public cloud. The platform uses managed relational databases, object storage, and containerized application services across a primary region. The board has approved a recovery time objective (RTO) of 4 hours and a recovery point objective (RPO) of 15 minutes for critical customer data. The CISO is concerned about both large-scale regional outages and ransomware affecting production workloads and backup repositories. Which strategy BEST meets the business requirements while aligning with sound cloud backup and disaster recovery practices?
- A
Use the cloud provider's standard in-region snapshots for databases and storage, and rely on infrastructure-as-code templates to rebuild the environment in the same region after an outage.
- B
Implement cross-region replication for critical data, maintain immutable or write-once backup copies in a logically separate account or subscription, and regularly test restoration of databases and application dependencies using documented runbooks.
- C
Configure nightly full backups of all systems to a single centralized backup vault in the primary region and prioritize anti-malware controls to reduce the likelihood of ransomware encrypting production systems.
- D
Deploy the application across multiple availability zones in the primary region and use auto-scaling to improve resilience, while accepting short-term data loss during regional recovery because backups can be reconstructed from application logs.
Show answer and explanation
Correct answer: B
Explanation
The key leadership decision in cloud backup and disaster recovery is to align architecture with business-defined RTO and RPO while addressing modern threats such as ransomware and control-plane compromise. For a 15-minute RPO, near-real-time or frequent replication for critical data is generally required; nightly backups are inadequate. For a 4-hour RTO in the event of a regional failure, recovery capability must exist outside the primary region, supported by documented and tested procedures. Security best practices also emphasize protecting backup integrity through immutability, write-once retention, and logical separation such as a different account, subscription, or backup security boundary. This reflects widely accepted guidance from cloud providers and security frameworks: backups should be geographically resilient where required, access to backup management should be tightly controlled, and organizations should regularly test restoration rather than assume backups are recoverable. From a CCISO perspective, the best answer balances business continuity targets, cyber resilience, governance, and operational execution.
- A. Incorrect.
This is insufficient for the stated requirements. In-region snapshots may help with localized failures, but they do not adequately address a regional outage. Rebuilding only in the same region fails the disaster recovery requirement for region-wide disruption. In addition, standard snapshots without immutability or separation may be vulnerable to deletion or tampering in a ransomware or privileged account compromise scenario.
- B. Correct.
This is the best answer. Cross-region replication supports regional disaster recovery and helps meet a 4-hour RTO and 15-minute RPO when designed appropriately for critical datasets. Immutable or write-once backup copies in a separate account or subscription reduce the risk that ransomware or a compromised administrative identity can alter or delete backup data. Regular restoration testing and runbooks are essential because backup success does not guarantee recoverability; tested recovery procedures are a core best practice in cloud resilience and business continuity planning.
- C. Incorrect.
This does not meet the RPO and regional resilience requirements well. Nightly full backups create a potential data loss window far larger than 15 minutes. Storing backups only in the primary region leaves the organization exposed to regional outage risk. Anti-malware is important, but it is not a substitute for resilient backup architecture, immutability, and tested disaster recovery capabilities.
- D. Incorrect.
Multi-availability-zone deployment improves high availability within a region, but it is not the same as disaster recovery for a regional outage. The option also explicitly accepts data loss, which conflicts with the approved 15-minute RPO. Reconstructing from application logs may be incomplete, operationally risky, and unlikely to satisfy governance expectations for critical customer data recovery.