Graph Database Applications in Financial Technology: Architecture, Challenges, and Ant Group’s GeaBase

Speaker and Context The session, presented by senior technical expert Fu Zhisong from Ant Group and edited by Hoh, focuses on the application of graph databases in the financial sector.

Fintech Architecture Evolution Three stages are described: (1) Business digitization – moving offline services online; (2) Data‑driven – using accumulated data for decision‑making, leading to services like Alipay, Yu’ebao, and Huabei; (3) Data intelligence – mining large volumes of data to discover new business models and enable real‑time, intelligent decisions.

Challenges in the Data‑Intelligence Era Ant Group faces increasing business variety, massive data volumes (over 100 billion records daily, peak transaction rates exceeding 400 k per second), complex relational computations, and the need for rapid, real‑time decision making.

Technical Response: Storage and Compute Evolution Two dimensions of upgrade are required: (1) Storage structure – from traditional relational databases to KV stores, NoSQL, and finally multi‑dimensional graph storage with temporal capabilities; (2) Compute paradigm – moving from offline and near‑real‑time processing to fully real‑time graph computation, reducing interaction cycles between compute and storage nodes.

Ant Group’s Graph Storage Practice GeaBase (Graph Exploration and Analytics Database) is a large‑scale distributed real‑time graph database designed for financial scenarios. It provides high‑performance query and analysis, multi‑dimensional temporal graph queries, and supports languages such as GQL and Gremlin. Its architecture separates storage and compute, allowing direct communication between storage nodes to minimize latency and maximize throughput.

Application Cases • Social: Ant Forest processes billions of edges with millisecond‑level response times. • Financial: Real‑time fraud detection, credit‑card cash‑out prevention, and instant loan approval (the “310” model) rely on rapid graph updates and analysis.

Standardization Efforts Ant Group collaborates with ISO/IEC on Graph Query Language, contributes to national standards for graph database requirements, and co‑authors industry white‑papers and testing methods to promote a unified graph ecosystem.

Future Outlook Graph technology is expected to handle larger scales, improve security and compliance, accelerate decision‑making, and become the foundational layer for fintech innovation.

Q&A Highlights Answers cover the definition of temporal graphs, their use in real‑time risk control, GeaBase’s support for temporal graphs and large‑scale deployments, and its specific optimizations for financial stability and consistency.