Persistent successfully replaced a leading bank’s traditional Message Queue (MQ) system with Confluent’s Kafka-based streaming platform, delivering a fault-tolerant, scalable, and real-time data processing solution. The bank required a modern system capable of handling high-volume, high-velocity transactions with near real-time latency while ensuring seamless integration with its Core Banking System (CBS). By leveraging Confluent’s capabilities, Persistent enabled high throughput, low latency, and cross-region data availability. The transformation resultedin reduced downtime, streamlined migration from legacy systems, and an improved customer experience. Additionally, the bank now has the flexibility to leverage advanced Confluent features such as KStream and ksqlDB for future innovation.
Highlights
- Managing million+ transactions on daily basis
- Downtime reduced significantly
- Add-on capabilities for scale up operations in future
The Story
Challenges
The bank needed to replace its traditional MQ system with a modern, fault-tolerant platform capable of handling high data volumes with near real-time processing. The new system had to be scalable, support data replay, enable data sharing across regions, and integrate seamlessly with the existing CBS.
Solution Approach Used
To address these challenges, Persistent implemented Confluent’s data streaming platform with a combination of advanced integration and data governance strategies:
High-Throughput, Low-Latency Data Processing
- Confluent Platform was deployed to ensure real-time data streaming, reducing transaction processing delays and improving system responsiveness.
- Kafka’s distributed architecture enabled the bank to process millions of transactions perday with fault tolerance and scalability.
Cross-Region Data Availability
- Confluent Cluster Linking was utilized to enable data replication across multiple Kafka clusters in different geographies.
- This ensured real-time data availability and consistency across regional banking operations.
Seamless System Integration
- Custom containerized source applications were developed to connect the existing banking infrastructure with Kafka.
- These applications ensured smooth data ingestion, transformation, and forwarding without disrupting core banking operations.
Future-Ready Infrastructure
- The implementation was designed with an extensible archi., allowing the bank to leverage Confluent’s KStream and ksqlDB for advanced real-time analytics and event-driven processing.
- This positioned the bank for future innovations such as real-time fraud detection, personalized financial insights, and AI-driven decision-making.
Schema Governance and Data Consistency
- Schema Registry was implemented to enforce strict data governance, ensuring compatibility between applications consuming Kafka data.
- This minimized data inconsistencies and enabled smooth upgrades or modifications in the future.
Technologies Used
- Kafka-Confluent
- Oracle Banking Platform
- Oracle FLEXCUBE
- Kubernetes (K8)
Outcome and Benefits
- Scalability & Performance: The system now supports high-volume transactions with near real-time processing.
- Seamless Migration: The transition from the traditional MQ platform was completed with minimal operational disruptions.
- Improved Resilience: Downtime was significantly reduced, ensuring high availability.
- Future-Proofing: The bank now has the flexibility to adopt additional Confluent capabilities for enhanced data processing and analytics.
- Enhanced Customer Experience: Faster data processing has improved responsiveness, leading to better banking experiences for customers.
