Jiangsu Rural Credit Union Audit System — High Performance, High Stability

lHybrid Architecture: GBase 8a MPP Cluster integrates with DB2, each leveraging their strengths to maximize platform performance.

lMassive Data Management: GBase 8a MPP Cluster significantly expands data management capacity, scaling to hundreds of terabytes.

lHigh Availability: GBase 8a MPP Cluster ensures 24/7 stable system operation with no single point of failure.

The audit system data platform of Jiangsu Rural Credit Union is built using GBase 8a MPP Cluster and IBM DB2. The GBase 8a MPP Cluster manages business data tables for massive data querying and correlation. The cluster comprises 14 nodes, with every two nodes forming a security group. The DB2 database stores system tables and handles transactional operations for the application systems, running on two minicomputers configured as an active-standby pair.

The audit system of Jiangsu Rural Credit Union requires managing data over a 3+1 year retention period, storing up to 30TB of data with a daily increment of 100GB, and completing data loading within a one-hour window.

During business hours, the system must support 100 concurrent users performing audit queries. Overnight, it handles over 6,000 batch jobs, and the maximum execution time for audit models must be under 10 minutes.

The audit system of Jiangsu Rural Credit Union (JSRCU) centrally collects and stores data from key business systems—such as core banking, credit, and financial management—as well as external data from sources like business registration, taxation, and utilities. It provides a convenient and efficient platform for data query, verification, and comprehensive analysis. By continuously and systematically analyzing both internal and external data, the system promptly detects anomalies in audited entities, assesses unusual business operations, and delivers targeted corrective measures and recommendations—all contributing to higher audit quality management.