GBASE supports the Bank of Tianjin Credit Card Data Real-Time Query Project.

As Bank of Tianjin advances its digital transformation strategy, data plays an increasingly decisive role in business analytics and internal decision-making. To establish a centralized Analytics Center and fully support the development of its unified reporting platform, the bank plans to deploy a distributed analytical database offering high-performance parallel computing, high availability, active-active deployment, and exceptional scalability. Working in tandem with Bank of Tianjin’s existing big data platform, this database will serve as the foundation for a financial big data analytics and service platform. By standardizing data and business rules, the bank aims to build business and data mid‑platforms, creating a big data platform that is high‑performance, highly elastic, highly secure, and independently controllable. This will significantly enhance data processing, scalability, and business support capabilities, ultimately achieving the IT strategic goals of a “Stable and Secure Bank” and a “Smart Bank”.

This project starts with Bank of Tianjin’s credit card real‑time data query system, re-architecting it by replacing the existing ADS/RDS databases with GBase 8a MPP Cluster from General Data Technology (GBase). Through a pilot deployment and a smooth transition, application capabilities will be gradually expanded, fully leveraging the storage and computing power of GBase 8a MPP Cluster to meet the demands of Bank of Tianjin’s unified reporting platform in data management, model computing, and business analytics.

The overall data architecture of Bank of Tianjin is outlined below:

The GBase 8a MPP Cluster database interfaces with the Bank of Tianjin's big data platform, supporting data synchronization via existing tools such as DataWorks and DataX to replace the original ADS/RDS databases. The data flow is as follows:

Based on the bank's overall plan, this project is designed to deploy a 3-node cluster for development and testing environments, and a 3+3 primary-standby cluster for the production environment—a "3+3+3" deployment model. For this phase, the production and test clusters will be built first, with the disaster recovery cluster scheduled for completion in 2023.

As of September 2023, the GBase 8a MPP Cluster has expanded its application scope beyond its original design, encompassing more business scenarios.

• The system supports comprehensive querying and analysis for business applications, significantly improving efficiency; complex queries on tables with tens of millions of rows deliver sub-second response times.

• Leveraging the GBase 8a cluster’s high scalability and performance, it provides a massive data storage and processing platform, enabling customers to support data volumes of hundreds of terabytes.

• Efficient technical support and a comprehensive training system build strong technical readiness for the customer in extending data warehouse applications and undertaking architecture upgrades.