Shanghai Telecom MPP Data Platform

Shanghai Telecom's ODS/EDW foundational data and application systems face the following issues:

Insufficient database storage;

ETL performance bottlenecks. Adding applications later may lead to resource contention, degrading overall program performance;

To address these challenges, a distributed data processing platform was built as a valuable complement to the ODS/EDW data warehouse. The GBase 8a MPP Cluster database, with its massively parallel processing architecture, delivers efficient processing power and strong join query performance. It breaks through the performance bottlenecks of loading high-volume business data, overcomes the shortcomings of traditional relational databases—such as high costs, long computation times, and delays in analytical results and business reports—and adopts a commodity PC server model, significantly reducing new storage acquisition costs.

Accelerated Operations: Meet the group’s mandate for faster enterprise-wide data operations—ensuring all product instances, sales offerings, and full datasets are uploaded before 10:00 AM every day.

Ad Hoc Queries: Self-service report queries now respond within seconds, resolving sluggish performance for high-volume data applications and noticeably improving user experience and satisfaction.

Performance Boost: Harness the high processing power and robust join query capabilities of a massively parallel processing (MPP) architecture to break through data-loading bottlenecks for large-scale business data. This overcomes the high cost, long runtime, and delayed analytics and business reports of traditional relational databases.

High Scalability: Move from a legacy vertical scale-up model to a horizontal scale-out approach driven by data volume, enabling online linear scaling where performance grows linearly with each additional node.