The bank's existing ODS-supported, decentralized data architecture could no longer meet operational requirements, presenting four key challenges:

1.Short Data Retention

13-month online data retention proved insufficient for historical analysis needs.

2.Limited Data Services

Batch-based delivery struggled with:

• High-concurrency queries

• Precision single-record retrieval

3.Complex Metric Processing

Multi-layered indicator calculation required excessive resources for:

• Data lineage tracing

• Logic verification

4.Fragmented Data Assets

Isolated systems and scattered storage hindered enterprise-wide data utilization.

Upgrade the current data system service architecture, which is supported by ODS and data warehouses, to a Data Lake–driven architecture.

This transition lays a solid technical foundation for the seamless implementation of future engineering projects and business modeling initiatives, enabling the integration of Data Lake and Data Warehouse (Lakehouse Architecture).

By fostering mutual reinforcement between systems, it aims to build a Data Lake with characteristics tailored to the specific needs of the city commercial bank, while effectively addressing challenges encountered in data projects.

The project successfully achieved the overall goal of migrating the ODS system to the Big Data Platform's Data Lake, integrating standardization, normalization, and process optimization. It has laid a solid foundation for unlocking the full value of the new data platform for the city commercial bank.

Additionally, the project has significantly enhanced overall data service capabilities and optimized the data ingestion process, greatly supporting data analytics and decision-making across the bank’s various business lines.