GBase 8s Centralized Transactional Database Dual-Site Three-Center Disaster Recovery Solution

Database systems are the core of any information system and a prerequisite for stable operations—demanding high security and reliability, especially in mission-critical business systems. A centralized database consolidates all data at a single center, where every application connects for reads and writes. While this model simplifies management and O&M, it introduces single-point-of-failure risks and performance bottlenecks that threaten system sustainability. To address these challenges, enterprises typically deploy centralized databases in a two-city, three-center architecture to ensure business continuity.

Deploy an SSC cluster at the primary site for load balancing and multi-node read/write services. The same-city disaster recovery site deploys HAC nodes in hot standby, ensuring rapid failover if the primary site goes down. Normally, the standby nodes are read-only, handling queries and reports to offload the primary system. A remote DR site deploys RHAC nodes, enabling disaster recovery over thousands of kilometers. It supports forced switchover to primary under manual intervention, or automatic failover upon primary equipment failure. All sites are managed through the CM connection manager, which routes service requests to specific cluster nodes based on SLA rules.

• With Hot Standby Capability

High-availability database clusters feature hot standby capability. In the event of a system failure, they can automatically switch to the standby node rapidly and restore database service functionality, ensuring business continuity and minimizing customer dissatisfaction caused by service interruptions.

• Data Consistency Between Primary and Standby Systems

After a customer completes a transaction, data is synchronized in real time across all nodes. If a database switchover occurs due to a failure at this point, the customer can continue using subsequent services seamlessly after the switchover. This avoids data inconsistency caused by the switchover and eliminates the need for customers to reprocess transactions.

• Automatic Switchover Process

The primary-standby switchover within the cluster is automatically performed by the database management system (DBMS) without manual intervention. The switchover is transparent to the business system, which requires no modifications. For example, if a service interruption occurs due to database hardware or software failures, the business system will automatically resume normal operation once the database is recovered.
 

• Sub-second Level Switchover Time

It ensures real-time data synchronization and consistency between the standby and primary nodes, supporting failure detection and automatic takeover of primary node services by the standby node. RTO (Recovery Time Objective) < 20 seconds, RPO (Recovery Point Objective) = 0.

• Remote Disaster Recovery Capability

If all database nodes in the production data center lose data service capabilities due to power outages, data center equipment damage, or other reasons, the high-availability database cluster can provide remote disaster recovery and switchover capabilities over distances of hundreds or thousands of kilometers, ensuring business continuity.