Definition
Region Failover refers to the automatic switching of traffic and services to a backup region when the primary region fails, ensuring business continuity. In the International IoT Cloud Platform (ICMP), Region Failover typically uses Multi-AZ (Multiple Availability Zones) or Multi-Region deployment, working together with Failover and Multi-Region to ensure 24/7 availability, meeting the high availability requirements for cross-border connection management.
Working Principle
The core of Region Failover is 'primary-standby switching': when the primary region is functioning normally, it handles all requests, while the standby region remains in hot standby or warm standby mode; when a failure in the primary region (such as data center power outage, network interruption, or natural disaster) is detected, the Failover mechanism is triggered, and DNS or load balancing switches traffic to the standby region. Data must be kept synchronized through cross-region replication, and a brief RPO/RTO may occur during the switch. Core components such as ICMP API Gateway, BSS, CDR processing, etc., must all support Region Failover to ensure that customer RESTful API calls, usage queries, and billing are unaffected.
Use Cases
In cross-border IoT scenarios, customer devices are spread worldwide, and the ICMP platform must operate stably 24/7. Region Failover guarantees: Activation and usage query—customers can activate SIMs and check usage at any time, unaffected by failures in a single region; Billing and Unified Invoice—CDR processing and the billing engine must remain continuously available; Alarms and Event Subscription—events such as device anomalies and Suspension must be pushed in real time. Combined with Geo-Distributed and Edge Location, it can further reduce cross-region latency and enhance the global customer experience.
Practice in ICMP
SimtoLink ICMP adopts a Multi-Region architecture, deploying production environments across multiple geographical regions to support Region Failover. Core services (such as API Gateway, BSS, and usage aggregation) have cross-region replication and automatic failover capabilities, with RTO (Recovery Time Objective) controlled at the minute level. When customers access via RESTful API or Global Dashboard, they do not need to perceive the underlying region switching. The platform regularly conducts failover drills to verify switching processes and data consistency, ensuring rapid recovery during actual failures.