Massive Legacy Resource Migration for Mobile Cloud
As cloud architectures mature, service providers often face the complex challenge of migrating vast amounts of data from legacy environments to modern infrastructure without disrupting active services. The following case illustrates how a major operator managed this transition at scale.
Client Needs and Challenges
As part of its early development phase, “Mobile Cloud” established resource pools in North China (Beijing), Central China (Changsha), and South China (Guangzhou). To enhance customer experience, improve service quality, and deliver a more stable, high-performance product, the operator decided to migrate services from these legacy pools to new, optimized infrastructure.
The scope of this migration was massive, involving over 4,000 enterprise clients, 100+ internal proprietary services, and a total of over 14,000 cloud host instances.
Implementation Solution
This migration encompassed a diverse technical landscape, including ECS hosts, RDS databases, OSS object storage, microservices, IoT platforms, and various standard and non-standard database types. The project faced high technical complexity due to the sheer volume of hosts, the scale of the customer base, and the critical requirement for minimal downtime.
To address these challenges, Information2 provided a comprehensive, 1-on-1 professional migration service for every Mobile Cloud tenant. This end-to-end delivery model covered client coordination, infrastructure surveying, resource provisioning, solution planning, execution, and final cutover. By utilizing real-time data replication software, system live migration tools, and specialized database migration utilities, the team achieved a secure, comprehensive migration of all data without disrupting active services.
Client Benefits
By leveraging Information2’s professional online migration software and technical expertise, the operator significantly reduced the technical complexity associated with migrating legacy resource pools. The solution improved migration efficiency and success rates, leading to a substantial increase in tenant satisfaction and accelerating the overall business development of Mobile Cloud.
Long-Distance Disaster Recovery for Regional Carrier Core Systems
For telecommunications operators, maintaining the availability of core billing and support systems is non-negotiable. This requires a disaster recovery architecture capable of spanning vast geographic distances to protect against regional disruptions.
Client Needs and Challenges
The “XX System” (a core billing and support system) is the heartbeat of this operator’s business, requiring strict 7×24 High Availability (HA). The primary requirement was the ability to rapidly failover to a standby database in the event of a primary database failure, ensuring service continuity.
Crucially, to meet regulatory and safety standards for disaster recovery, the backup center had to be located more than 300 kilometers away from the production site. Furthermore, the operator required a mechanism for regular disaster recovery drills to validate the reliability of the solution. These drills needed to simulate primary database failures to verify that the switchover and data synchronization processes were functioning correctly. This required the establishment of two independent MySQL clusters; however, traditional synchronization methods struggle to maintain sync relationships between independent, long-distance clusters.
Implementation Solution
Information2 implemented a solution based on MySQL cross-cluster semantic replication, utilizing the GTID (Global Transaction Identifier) mechanism. GTID is essential for ensuring data consistency and integrity between distinct MySQL clusters.
In this replication architecture, the GTID mechanism tracks transaction history, allowing for data checkpoint resumption. If a network interruption or switchover occurs, the system can resume transmission exactly where it left off. This capability significantly enhances the stability of synchronization, even over long distances.
Client Benefits
Deploying two independent MySQL clusters for disaster recovery delivered multiple strategic advantages for the mobile billing system:
- High Availability: The architecture ensures that if the primary system fails, the secondary cluster takes over immediately, minimizing downtime.
- Data Reliability: Robust replication ensures data is protected and can be restored even in the event of corruption at the primary site.
- Performance and Scalability: The architecture supports load balancing, reducing pressure on the primary system and optimizing the user experience.
- Validated Resilience: The ability to perform regular DR drills verifies the system’s rapid recovery capabilities.
Ultimately, this dual-cluster approach guarantees business continuity. Whether facing hardware failure or unexpected outages, the operator can maintain critical billing and communication services, ensuring system stability and high user satisfaction.