GBASE Empowers Shandong Mobile's Anti-Fraud Big Data Service

The anti-fraud big data business has long been tightly coupled with the big data platform's business analytics operations, leading to resource contention between services and global business impact when hardware failures occur. This not only compromises the real-time requirements of the anti-fraud business but also frequently causes timeouts against the group's assessment criteria. To further decouple services, modernize the architecture, and improve the real-time performance of the anti-fraud big data business, a dedicated anti-fraud big data platform must be built.

Based on the characteristics of the anti-fraud big data business and future scalability, this project leverages GBase 8a MPP V95 for support. The management nodes and compute nodes adopt a federated architecture to deliver enhanced high concurrency. Considering data volume, a primary-secondary dual-replica configuration ensures data redundancy and high availability, meeting both real-time and scalability requirements.

1) Enhanced Storage and Computing Power. GBase 8a MPP provides structured, massive-scale storage and computing capabilities to meet future business growth needs.

2) High Data Availability. A data replica redundancy mechanism mitigates the instability of x86 hardware, delivering data security for anti-fraud operations.

3) Real-time Performance. A single cluster ensures the independence of decoupled services and improves the timeliness of batch data processing.

4) Fine-grained Business Operations. Service decoupling is the inevitable direction for future big data development, as it prevents resource contention between mixed workloads and avoids system-wide failures.