When an application is designed that requires high availability, what is a reason to use a cross-region cloud?

A cross-region cloud architecture involves deploying an application across multiple, geographically separate regions. The primary purpose of this design is to provide disaster recovery (DR). A disaster, such as an earthquake, hurricane, or widespread power outage, could potentially disable an entire cloud region. By having a replica of the application and data in a different region, services can be failed over, ensuring business continuity. This strategy protects against large-scale, regional failures that a multi-zone deployment within a single region cannot.

B. Protect from a single component failure: This level of failure (e.g., a single server or disk) is typically handled by redundancy within a single Availability Zone, such as using multiple instances or redundant storage.

C. Minimize costs: A cross-region deployment significantly increases costs due to duplicated infrastructure, management overhead, and inter-region data transfer fees. It is a strategy for resilience, not cost optimization.

D. Account for failure in another zone: This describes a multi-Availability Zone (multi-AZ) architecture within a single region. It provides high availability against the failure of a single data center (zone), not an entire region.

1. AWS Well-Architected Framework

Reliability Pillar (July 31

2023): In the section "REL 6: How do you plan for disaster recovery (DR)?"

the best practice REL06-BP01 states

"To improve availability and reduce latency

you can deploy your workload to multiple AWS Regions. This architecture can be used for disaster recovery (DR) by providing a geographically separate location to which you can fail over."

2. Microsoft Azure Documentation

"Business continuity and disaster recovery (BCDR): Azure Paired Regions" (Updated: 09/18/2023): The document states

"A primary goal of Azure paired regions is to provide for disaster recovery (DR)... This design choice minimizes the likelihood of natural disasters

civil unrest

power outages

or physical network outages affecting both regions at once."

3. Google Cloud

"Disaster recovery planning guide" (Updated: 2024): In the section "Building blocks for disaster recovery

" the guide explains

"A multi-region architecture can provide protection against the loss of an entire region. If a natural disaster causes a whole region to go offline

your application can continue to serve traffic from another region." This explicitly links multi-region design to disaster recovery.