GBASE database empowers a commercial bank in Hainan to build a high-performance, standardized data platform.

In recent years, as interest rate liberalization accelerates and internet finance booms, traditional banks are being reshaped by the wave of digital disruption, giving rise to entirely new products, services, and business models. At the same time, big data technology has matured significantly, offering higher cost efficiency, stronger compute and storage capabilities, and seamless scalability. It unlocks the full value of massive data and has become a foundational must-have platform for banks to thrive in the era of internet finance and big data.

Against the backdrop of the accelerated development of the Hainan Free Trade Port, a provincial-level urban commercial bank in Hainan is entering an entirely new phase of business growth. Diverse business models are emerging rapidly, and data volumes are soaring. As business data grows richer in content and form, higher demands are placed on overall data support and service capabilities. It is therefore imperative to build a big data platform that integrates existing internal data with external data sources, and that boasts strong parallel processing power and easy scalability.

The bank adopted a hybrid architecture combining an MPP database cluster and a Hadoop cluster to build a low-cost, high-performance, high-capacity, and easily scalable foundational data platform. The MPP database cluster runs GBase 8a MPP Cluster from General Data Technology as its computing engine layer, delivering powerful analytical processing to the entire data platform.

A commercial bank in Hainan introduced GBase 8a MPP Cluster, the distributed logical data warehouse from General Data Technology, combined with a Hadoop cluster to build a low-cost, high-performance, large-capacity, and easily scalable data infrastructure platform. This enables centralized collection, storage, processing, analysis, and application of structured, semi-structured, and unstructured data. Based on this architecture, the bank rebuilt its data warehouse and standardized historical data storage and usage, forming a standardized, high-performance foundational data system that provides data support for the bank's operations management and business applications.

The unified data platform consolidates the bank’s primary business data. As part of the process, the data platform first processes and standardizes common data. On top of this data foundation, a flexible, self-service, open, easy-to-use, and secure data self-service portal was built, integrating data analysis and visualization tools and other resources, enabling data-consuming departments to efficiently achieve business goals such as querying and analysis. It also boosts system development efficiency and response speed, quickly supporting the data demands of upper-layer applications.

The GBase database serves as the computing engine layer of the data middle platform, used for processing various subject-oriented models and batch jobs, and providing external data access capabilities. During the project implementation, it demonstrated leading performance and architectural advantages:

• Easy scalability: Adopts a fully parallel MPP + Shared Nothing federated architecture to deliver horizontal capacity and performance scaling for the data platform.

• Data fusion: Exceptional integration capabilities enable transparent data interaction with the Hadoop data lake, building a lakehouse solution across the entire data lifecycle.

• High performance and maintenance-free: Utilizes high-performance, maintenance-free intelligent indexing technology, where index space accounts for no more than 1%. The intelligent index contains column statistics that effectively filter data during retrieval, significantly reducing disk I/O and dramatically improving query performance on massive datasets.

• Efficient resource management: Through flexible configuration of resource pools and resource usage plans, it achieves vertical resource isolation for different database users, controlling key resources and metrics such as CPU, memory, disk space, disk I/O, and concurrent task count. It provides comprehensive multi-tenancy capabilities.

Data Platform Technical Architecture Diagram

The data platform deploys over ten GBase database nodes, managing petabytes of bank-wide business data to support applications in external regulatory reporting, asset-liability management, business management, and risk management. Delivering in-depth big data analytics and integrated applications across multiple business domains and scenarios, it provides a high-performance data engine for business growth.

• Supporting Business Objectives

Unify data resource management, comprehensively elevate data service capabilities, and fully unlock data value. Meet the bank’s needs for customer marketing, risk control, operational management, and external compliance, while advancing data governance and enhancing data asset utilization.

• Low Cost and Full Technology Independence

Built on an entirely China-developed hardware and software stack, the platform significantly reduces upfront costs while ensuring infrastructure security and reliability. It mitigates external risks in operation, maintenance, and upgrades, safeguarding financial data security.

• Business Continuity

Leveraging a Shared-Nothing architecture, the platform supports online, on-demand scaling without service interruption, effectively ensuring continuous operations and providing a robust foundation for business innovation.

• Hybrid Architecture for Unified Data Aggregation

GBase databases enable transparent data interaction across heterogeneous platforms and, together with the Hadoop cluster, form a lakehouse architecture that covers the full data lifecycle.