Urumqi Bank ODS Data Application Netezza Replacement Project

Project Background

As banking institutions undergo digital transformation, data has become an increasingly vital asset, making enhanced data management capabilities more important than ever. Since 2014, Urumqi Bank has built data management and application information systems based on its ODS and data warehouse, extracting, storing, and processing data from various source systems. To meet the demands of digital transformation and regulatory compliance, and in line with the bank’s strategic development, Urumqi Bank plans to replace the database software of its ODS system.

Solution

By rebuilding the infrastructure and database of the ODS and data warehouse systems, the project resolves the scalability bottlenecks and overall performance limitations of the existing integrated appliance. The original IBM Netezza database is replaced with a China-developed distributed analytical database running on x86 servers, splitting the original database into three clusters: a primary database cluster, a historical data cluster, and an application service cluster. This ensures that the entire ODS system has no single point of failure, enables flexible expansion of each cluster, and guarantees continuous, stable, and high-performance data services.

The original ODS data architecture encompasses data ingestion, interface layer, integration layer, summary layer, application layer, and a data sharing platform. Building on this foundation, the rebuilt architecture consolidates into interface, integration, summary, data mart, and historical layers. GBase 8a MPP Cluster replaces IBM Netezza to power the interface, integration, summary, data mart, and historical layers, providing full control over end-to-end data flow management. The construction of application service marts reduces interference between processes and applications, thereby improving data service stability.

Application Results

First domestic database replacement in the financial industry of Xinjiang Uygur Autonomous Region

This project marks the first instance in the Xinjiang financial industry where an IBM Netezza system was replaced by GBase 8a MPP, achieving self-sufficiency and full control over the database layer of Urumqi Bank’s ODS.

Significantly enhanced analytical capabilities, empowering big data analysis and decision-making

The analytical database GBase 8a MPP is purpose-built for OLAP scenarios, performing analysis and modeling on massive datasets. It delivers high-performance processing for complex multi-table joins on large data volumes, enabling fast queries and analysis. It reveals data patterns and trends for upper-layer systems, effectively supporting decision-support systems.