"Design a multi-region active-active deployment architecture for a critical e-commerce application on Alibaba Cloud. How do you handle data synchronization and failover?"
Alibaba Cloud is a core business unit, and Alibaba itself runs a multi-region active-active architecture. This question assesses your understanding of cloud-native design, disaster recovery, and the unique challenges of keeping data consistent across geographically distributed systems.
Define the architecture: active-active vs active-passive, and why active-active is preferred for availability-critical applications.
Explain traffic routing: DNS-based geo-routing, GSLB (Global Server Load Balancing), with health-check-based failover.
Address data synchronization: unit-based architecture (单元化) where user data is sharded by geography, with cross-region async replication for global data.
Cover failover and testing: automated failover triggers, regular disaster recovery drills, and data consistency verification.
I'd design this using Alibaba's unit-based architecture (单元化) pattern. Users are assigned to a "unit" (home region) based on their geographic location or user ID hash. Each unit is self-contained with its own application servers, databases, and caches, capable of serving all requests for its users independently. For data synchronization, user-specific data stays in the home unit with async replication to a backup unit for disaster recovery. Global data like product catalogs uses a master-region model with sub-second replication via DTS (Data Transmission Service) to all regions. Traffic routing uses Alibaba Cloud's GSLB with health checks every 5 seconds — if a region's health score drops below threshold, traffic is automatically redirected to the backup unit. The failover process is: detect failure (health check), drain connections (30s grace period), switch DNS, and warm up caches in the target region. I'd run quarterly disaster recovery drills where we intentionally fail one region to validate the entire process end-to-end.
The 单元化 (unit-based) architecture is Alibaba's signature approach — mentioning it shows deep familiarity.
Know the difference between active-active and active-passive, and the CAP theorem implications of each.
Mention DTS (Data Transmission Service) — it's Alibaba Cloud's core data sync product.
It's Alibaba's approach to horizontal scaling across regions. Each "unit" is a self-contained deployment that handles a subset of users. It enables independent scaling, fault isolation, and simplified disaster recovery because each unit can operate autonomously.
For shared data like product info, use a designated master region for writes with async replication. For user data, route writes to the user's home unit. Conflict resolution uses last-write-wins with vector clocks for critical fields.
HireMe AI listens to your interview in real-time, detects questions, and generates instant answer suggestions — so you never freeze up again.
Try HireMe AI Free