GBASE supports Inner Mongolia Rural Credit Cooperative in building its big data platform.

After the merger of Inner Mongolia Rural Credit Cooperative (Inner Mongolia RCC) with city-level credit cooperatives, the organization faced the challenge of storing and managing massive volumes of business data. The total raw historical data has now approached 50 TB, and traditional data storage and management models are increasingly revealing bottlenecks in data processing and architectural scalability. To support future business growth, strengthen data resource integration, enhance data processing and analysis capabilities, improve data service quality and timeliness, and meet the demands of deep data analytics and big data applications, a new MPP database software product was needed to deliver more comprehensive, deeper, and more timely data services. Inner Mongolia RCC built a big data platform using a domestic data warehouse cluster product, obtaining a solution for handling and storing massive business data. The project has since been expanded into a dual-center architecture. The database cluster is now in stable operation, with an ongoing requirement to scale out cluster nodes.

The big data platform deploys a database cluster with 20 nodes, and data is protected with a three-replica scheme. It primarily supports the following business systems: the historical query system, the PBC regulatory reporting system, the reporting system, and the CRM system, undertaking the bank-wide daily incremental batch data processing tasks. Furthermore, starting in 2021, the bank required the big data platform to implement a dual-center model, enabling real-time data disaster recovery at the backup center.

The big data platform of Inner Mongolia Rural Credit Cooperative (IMRCC) is built on the GBase 8a MPP Cluster database from General Data Technology. The system uses a 20-node PC server cluster to form the data warehouse. Daily incremental data is loaded via file-based data transfer from source business systems such as management accounting, regulatory reporting, reporting platforms, and CRM. The data warehouse cluster now manages over 50 TB of raw data. Even at this scale, the system delivers high-speed batch data loading, processing, and querying. The platform has been deployed across dual data center sites in Horinger and Jinqiao, enabling automatic and real-time data synchronization between the primary and standby centers, fully meeting the cooperative's disaster recovery requirements.

Data Warehouse Architecture for Inner Mongolia Rural Credit Cooperative

1. Significantly accelerates the bank's daily batch data processing.

Under the same data volumes, the previous business platform experienced lengthy processing cycles. The current database cluster now completes related tasks within hours, fully satisfying data aggregation demands across relevant business scenarios.

2. Massive data storage capacity helps the bank effectively reduce hardware procurement costs.

The data warehouse cluster, which supports highly efficient data compression for the same data volumes, reduces hardware procurement costs compared to traditional storage models. The same physical storage space can accommodate and manage more business data.

3. Deployed an active-active model for the intracity disaster recovery cluster.

The active-active disaster recovery cluster effectively supports Inner Mongolia Rural Credit Union's dual data center management model. The active-active architecture of the GBase 8a database cluster robustly protects against data corruption and business disruption caused by unforeseen events such as power outages, fires, and flooding. It ensures uninterrupted access to business systems while maintaining data security and integrity.