Project Background
The Jinan Housing Provident Fund Management Center manages provident-fund accounts, loans, and withdrawal review services, and has long supported large volumes of public-service transactions.
As collection volume continued to grow, the center launched a next-generation platform under a unified management model, requiring a more scalable and resilient domestic database foundation.
Challenges and Requirements
- The original platform had been running for more than ten years and had accumulated a massive amount of historical data, including a single table with up to 170 million rows.
- The system had to serve tens of thousands of organizations and millions of employees, which required strong concurrency for account opening, closure, deposits, transfers, and loan services.
- Core business capabilities were split across many modules such as housing, loans, settlement, and auditing, creating complex data interaction and strong consistency requirements.
- Performance targets were demanding: average processing time under 3 seconds, annual interest settlement within 2 hours, and at least 320,000 tpmC under peak load.
Implementation Approach
Next-generation smart provident-fund platform
Highgo participated in the implementation of the new platform in line with the Ministry of Housing and Urban-Rural Development’s standards for large and medium-sized cities.
Multi-cluster database deployment
The platform uses internal core-business clusters, a certificate repository cluster, and an external core-business cluster, with Highgo managing all structured data across the environment.
Data fusion and metro disaster recovery
Highgo’s data-fusion platform synchronizes data to a metro DR center in real time, forming a resilient backup layer for key public-service systems.
Project Highlights
- Delivered more than 10x efficiency improvement in key scenarios such as loans and personal information queries.
- Introduced HA clustering to eliminate single points of failure in the data layer.
- Enabled real-time data transfer between the local data center and the metro DR center.
- Supported migration of nearly 500 GB of legacy data within two days with close engineering collaboration.
Customer Value
- Created a demonstrative domestic-database replacement case for digital government, public-sector localization, and finance-like citizen-service workloads.
- Completed de-IOE transformation by replacing DB2 and Oracle previously running on IBM minicomputers with Highgo on domestic platforms.
- Improved security readiness through multi-layer protection mechanisms that support compliance assessments and personal-data protection goals.
- Delivered stable production performance after go-live while preserving headroom for continued business growth.