Use Cases

Retail Banking

Banks rely on fast computer systems to process transactions and to manage customer interactions, and requirements for additional speed must be pursued by leveraging scalable, cost-effective IT solutions.

Consumers are increasingly turning to touchless and online interactions when dealing with their money. Web and mobile app usage continues to grow, which is fortunate for banks who rely on reducing costs by providing more self-service interfaces for customers. Non-cash payment systems are also enjoying more widespread use, and the added burden on transaction processing systems is leading more banks to modern, scalable systems. As online transactions continue to grow, especially fueled by the popularity of e-commerce, banks have to be especially cautious of fraud. Card-issuing banks need to ensure they are minimizing the financial risk of fraudulent transactions, while also ensuring that fraud detections are not too strict to cause “false positives” to miss out on transaction fees on legitimate activities.

Business Requirements

Retail banks face many well-known yet ongoing challenges around customer experience, responsiveness, reliability, and security, as they face new generations of customers with evolving usage patterns. These challenges drive business requirements that include:
  • Improving customer experiences via fast data access at scale with reads and writes in the millisecond range, to provide the performance to handle growth and spikes of online requests at the level of tens of thousands per second, without a costly overprovisioning of hardware
  • Taking action in real-time, especially to drive responsiveness that leads to increased revenue, such as timely promotions and offers
  • Adding new capabilities that create more customer-facing product offerings, with faster time-to-market, and without a costly system overhaul
  • Controlling existing expenses, especially mainframe MIPS costs, when expanding core banking functionality into new channels
  • Improving fraud detection techniques to reduce loss due to fraud while also retaining transaction fees that were previously lost due to a high rate of false positives
  • Boosting profitability by increasing IT agility and enabling faster development and deployment of internal business applications that focus on operational efficiency
  • Addressing all of the above while running a secure and highly available infrastructure
These are just some of the reasons that our banking customers choose us to run their critical systems.

Why Hazelcast

Hazelcast works with many retail banking customers who turn to us for speed at scale, security, and reliability. The Hazelcast Platform uniquely provides a distributed, in-memory data store combined with a high-speed stream processing engine, to run the fastest applications in any type of data-intensive environment. Consider some of the technology advantages that let Hazelcast customers run highly successful banking solutions:

Simplified development

Hazelcast was designed to simplify the application development process by providing a familiar, common-sense API that abstracts away the complexity of running a distributed application across multiple nodes in a cluster, allowing developers to spend more time on business logic and no time on how to write code to distribute compute work across available resources. Its cloud-native architecture requires no special coding expertise to get the elasticity to scale up or down to meet highly fluctuating workload demands.

Microsecond performance

Whether you are running large-scale transformations on data, running a digital integration hub (DIH) as a data layer to a costly backend mainframe, or running real-time calculations that make fast automated decisions, Hazelcast is designed for the ultra-performance that today’s banking workloads require. The proven performance advantage is especially valuable for data-focused experimentation that enables ongoing business optimization, especially in data science initiatives including machine learning inference for fraud detection.

Mission-critical reliability

With built-in redundancy to protect against node failures, and efficient WAN Replication to safeguard against total site failures, Hazelcast was built to provide the high availability to run mission-critical systems. The extensive built-in security framework ensures data is protected from unauthorized viewers, and security APIs allow custom security controls to be added more the most sensitive environments.

Use Cases

Building web and mobile apps that connect directly to core banking backend systems is expensive and time-consuming, especially since many backends were not designed for the modern interfaces of today. By deploying a digital integration hub (DIH) based on Hazelcast, banks can deliver a common API that delivers core banking data with in-memory speeds to new customer channels.

Mainframes still run many banking operations today, but extending these systems to do more often comes with much higher MIPS cost. By leveraging Hazelcast in a mainframe integration effort, much of the workloads that do not have to run on the mainframe can be handled in a more cost-effective way while retaining similar performance and availability characteristics.

Payment processing often entails a series of computations to determine the validity of a payment request, and the optimal routing of a transaction across payment partners. With Hazelcast, banks can run complex operations including sophisticated, machine learning-based fraud algorithms to reduce fraud loss while also reducing other operational costs.

Many banks have data sets that need to be shared/cached across different divisions, and integration with the systems of record often creates an excessive load that can cause disruptions. A distributed data-as-a-service on Hazelcast allows data sharing across different divisions and business applications, in a high-speed, cost-effective manner. And as much more than just a cache, Hazelcast can support other compute-intensive operations on shared data to run transformations, data enrichment, aggregations, and any other operations that makes the data more consumable for the various end-users.