Solution Overview

The fintech sector is rapidly transforming, driven by technological innovation, evolving customer needs, and regulatory changes. This dynamic landscape creates both opportunities and challenges for financial services providers:

• A high volume of real-time transactions requiring millisecond-level processing
• Complex data relationships spanning customer profiles, financial instruments, and privacy issues
• Strict regulatory compliance requirements with geographical data residency concerns
• Need for scalable and flexible systems that can adapt to spikes in volume and new product offerings
• Real-time analytics capabilities for risk assessment and fraud detection

We present a reference architecture using Hasura DDN (Data Delivery Network) with MongoDB Atlas to address these challenges. This combination offers a powerful, flexible, and scalable solution for modern fintech applications.

Hasura DDN is a universal data access layer for next-gen apps and AI. It lets teams effortlessly build and deploy a fast, secure, and federated API layer on all their data.

MongoDB Atlas is a leading fully managed cloud database service known for its flexible document model, horizontal scalability through sharding, and performance optimizations that support the demanding requirements of financial applications. Together, they create a foundation that accelerates innovation while maintaining the security and reliability essential to financial services.

Reference Architectures

The Hasura DDN with MongoDB reference architecture provides a comprehensive framework for building modern fintech applications that can handle high transaction volumes while maintaining data integrity, security, and regulatory compliance.

The diagram below illustrates the multi-regional deployment of Hasura DDN with MongoDB clusters, showing how fintech client applications connect through a global load balancer to distributed data nodes with centralized authentication, analytics, compliance, and external service integration.

This architecture diagram shows the key components and their interactions within the reference implementation. Fintech client applications connect through a global load balancer to multiple Hasura DDN regions, which in turn interact with MongoDB clusters deployed across different geographical regions. The architecture also incorporates centralized authentication and access control, AI and analytics capabilities, global compliance and security measures, and integration with external financial services.

Key components of the architecture:

• Data layer: MongoDB clusters for primary data storage
• API and access layer: Hasura DDN for data access and real-time subscriptions
• Application layer: Fintech services and applications
• Security layer: Authentication and authorization services
• Analytics layer: Data processing and machine learning services

Data Model Approach

The MongoDB data model for this fintech architecture leverages the document-based structure to address the unique requirements of financial applications. The flexible schema design allows fintech organizations to adapt quickly to new financial products or regulatory requirements without disruptive schema migrations. For optimal performance, the model employs strategic decisions around data organization.

Code Snippet

This model implements key approaches mentioned in the document: using MongoDB's flexible schema for financial product adaptation, supporting sharding strategies for high-volume data (potentially sharding by customer ID or date ranges for time-series transaction data), and leveraging MongoDB's storage capabilities for performance optimization. The structure also facilitates the role-based access control (RBAC) and field-level security mentioned in the Security and Governance section.

Building the Solution

The implementation of the Hasura DDN with MongoDB architecture for fintech applications follows a strategic approach focused on security, performance, and scalability. Building this solution requires careful consideration of how the MongoDB data layer interacts with the Hasura DDN API layer while ensuring that authentication, security policies, and compliance requirements are properly addressed.

The solution should be deployed in multiple geographic regions to support global financial operations, with MongoDB clusters configured for high availability using replica sets and appropriate sharding strategies based on financial data access patterns. Hasura DDN instances should be positioned close to their respective MongoDB clusters to minimize latency for real-time financial transactions and market data updates. The centralized authentication and access control layer ensures consistent policy enforcement across all regions, while the AI and analytics layer enables advanced capabilities like fraud detection and risk assessment.

Organizations implementing this architecture should adopt an incremental approach, starting with specific financial use cases like trading platforms or personalized banking, and gradually expanding to cover more complex scenarios. The solution can accommodate both new fintech startups building from scratch and established financial institutions transitioning from legacy systems through the API-first modernization approach described in the reference architecture.

For those interested in exploring anti-money laundering (AML) use cases more in depth, we recommend checking out the Axiom repository, which provides a comprehensive demonstration of implementing AML solutions using this architecture.

Note: While the API querying functionality will work as documented, setting up PromptQL locally requires additional steps not covered in the repository. You can access the PromptQL playground via this link: https://promptql.console.hasura.io/public/aml/playground

Key Learnings

Implementing the Hasura DDN with MongoDB architecture for fintech applications reveals several important insights that organizations should consider:

1. Distributed data architecture drives performance: The combination of MongoDB's distributed clusters with Hasura DDN's regional deployment model enables financial institutions to achieve the microsecond-level latency required for trading platforms and high-frequency transaction processing while maintaining data consistency across global operations.
2. Security must be multi-layered: Effective security in fintech applications requires both centralized and decentralized approaches. The hybrid policy management strategy leveraging Hasura's permission systems and MongoDB's field-level security provides comprehensive protection while maintaining flexibility for evolving regulatory requirements like GDPR, PSD2, and MiFID II.
3. Data model flexibility accelerates innovation: MongoDB's schema flexibility paired with Hasura's GraphQL API generation capabilities allows fintech organizations to rapidly introduce new financial products and adapt to regulatory changes without extensive redevelopment, significantly reducing time-to-market for new offerings.
4. Real-time capabilities transform customer experiences: The architecture's support for real-time data subscriptions and complex relationship mapping enables next-generation financial applications like personalized banking, immediate fraud detection, and comprehensive risk assessment that leverage a complete view of customer data.
5. Modernization can be incremental: The API-first approach allows organizations to gradually transition from legacy systems by creating a modern data access layer while maintaining existing data sources, reducing risk during digital transformation initiatives in the highly regulated financial services industry.

Technologies and Products Used

MongoDB:

• MongoDB Atlas

Partner technologies:

• Hasura
• Apache Kafka for event streaming

Authors

• Jon Mills, Hasura
• Aditi Phadke, Hasura
• Asawari Samant, Hasura
• Adam Malone, Hasura
• Kenneth Stott, Hasura
• Sebastian Rojas Arbulu, MongoDB