Download PDF
Case Studies > Hazelcast Powers Real-Time Fraud Detection

Hazelcast Powers Real-Time Fraud Detection

Technology Category
  • Analytics & Modeling - Real Time Analytics
  • Application Infrastructure & Middleware - Data Exchange & Integration
  • Application Infrastructure & Middleware - Database Management & Storage
Applicable Industries
  • Finance & Insurance
Applicable Functions
  • Business Operation
Use Cases
  • Fraud Detection
  • Real-Time Location System (RTLS)
Services
  • Software Design & Engineering Services
  • System Integration
The Challenge
Competent fraud detection systems are crucial for organizations to identify financial crimes such as payment card fraud, anti-money laundering, and anti-terrorist financing. The increasing sophistication of fraudsters' systems and the growing volume of financial transactions put immense pressure on existing detection systems. In the UK, payment card spending exceeds £180 billion from 3.9 billion transactions, with financial fraud losses consistently rising. Detection systems must be flexible, report anomalies in near real-time, and be available 24/7 to cope with these challenges.
About The Customer
A top ten US bank faced significant challenges with its existing fraud detection system, which was built on an old relational database platform. The system's limitations caused the bank to hit transaction rate limits, breaking fraud detection SLAs and hindering new business opportunities. The bank needed a more efficient and scalable solution to handle its growing data and transaction volumes, ensuring real-time fraud detection and compliance with SLAs.
The Solution
Hazelcast In-Memory Data Grid (IMDG) and Hazelcast Jet were implemented to address the bank's challenges. Hazelcast IMDG stores account details, recent transaction history, and scoring in one place, enabling powerful and comprehensive checks within short SLAs of 500ms. The in-memory technology allows for fast data ingestion and transformation, while the distributed architecture ensures high availability and scalability. Hazelcast Jet provides in-memory stream and fast batch processing, further enhancing the system's real-time capabilities. The bank also uses WAN replication to synchronize multiple Hazelcast clusters, ensuring service continuity in case of regional data center outages.
Operational Impact
  • Hazelcast's in-memory technology allows the bank to process 5,000 transactions per second, with expectations to grow to 10,000 transactions.
  • The distributed architecture of Hazelcast ensures high availability and scalability, allowing the system to handle large volumes of data and transactions without performance degradation.
  • The implementation of Hazelcast has enabled the bank to meet its fraud detection SLAs, ensuring real-time detection and compliance.
Quantitative Benefit
  • Hazelcast is used to store 2TB of customer data, with plans to grow to 5TB.
  • The bank can process 5,000 transactions per second, with expectations to grow to 10,000 transactions.

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.