Use Cases

Open Banking

Learn how in-memory technologies can satisfy the EU Revised Payment Service Directive (PSD2) in support of Open Banking.

Why Open Banking?

The European Union “Revised Payment Service Directive” (PSD2) brings banking and financial institutions closer to enable bank customers, both consumers and businesses, to use third-party providers to manage their finances. Third-parties have access to bank data and customer data to create value-added services such as secured and unsecured loans, mortgages, etc. Welcome to the age of Open Banking; it will touch our day-to-day lives by providing services tailored to individual needs in purchasing, financial services, travel, health insurance, financial technology, and investing.

All of these opportunities will require:


Open Banking requires a different level of speed due to more complex back-end integration requirements across a broader range of services. This is where the speed of in-memory can ensure you thrive.


Open Banking exposes what had been a closed system to a broad array of service providers, all of whom require not just speed, but speed on a much larger and more complex scale. The Hazelcast distributed processing architecture can help you stay one step ahead of a fast-moving demand curve.


The challenge with Open Banking is that if something fails on the back-end, it doesn’t just affect your bank, it impacts your entire payment ecosystem. A cloud-based distributed architecture provides instant redundancy, so system failures become a low probability event.


A groundswell of new companies accessing your back-end systems means more opportunity for fraud. Hazelcast enables fraud detection algorithms that easily exceed even the most stringent SLAs. This specific use case is one of our core competencies.

The optimal solution to the complexities of Open Banking

The Hazelcast Platform delivers the necessary ingredients for success such as performance under higher workload, scalability both horizontally and vertically, support for analytic and transactional processing, cross-application programming language support and more. The platform natively integrates in-memory computing and stream processing to deliver elastically scalable Open APIs with ease.

Enable excellence and innovation across your banking ecosystem

Exceed expectations

Open Banking introduces substantial risk as well as opportunity across your partner and customer ecosystem. In-memory technology gives you the adaptability and speed needed to exceed performance and security SLAs under the most demanding conditions.

Meet burst requirements

Burst requirements in Open Banking are going to have much broader data sources and more integration complexity, while expectations will continue to climb. This is an ideal scenario for the speed, stability, and security enabled by distributed in-memory computing.

Improve performance

Hazelcast is delivering industry-leading performance at scale for some of the most demanding banks and payment processors in the world, while driving core enablement capabilities against a broad, complex, and demanding ecosystem.

Aggregations per second
Across multiple networks and transaction types

Security in Complex Environments
10K TPS w/SSL/TLS 1.2 w/Open SSL
Consistent security across a broad and dynamic payment ecosystem

Hazelcast enables the transition to Open Banking

The Hazelcast Platform has an in-memory data store at its core to provide the fastest access to data. This allows a bank to hold all or active data sets in memory while integrating with disk-based stores like RDBMS, Hadoop, MongoDB, etc. Hazelcast also provides a stream and batch processing engine which uses the in-memory data store to provide high-speed event processing with common functionality such as data lookups, enrichment, transformation, analytics, etc. Hazelcast models your workflows as directed acyclic graphs (DAG) to optimize your application logic for you — cutting down on setup and integration time.

The Hazelcast Platform for Open Banking key features

  • Platform with integrated processing and data storage
  • Supports real-time and batch event processing
  • Temporal sequencing of events
  • Supports advanced analytical operations
  • SQL-based analytics for both data-at-rest and data-in-motion
  • Supports multiple programming language clients: Java, C++, C#/.NET, Node.js, Golang, Python, Scala
  • Scale-up and scale-out of system resources
  • Ready-made connectors to consume events from various data sources
  • Easy to embed or integrate based on industry-standard APIs, which is critical to Open Banking

The Hazelcast Platform is very easy to integrate with other applications and services. It can be integrated with middleware like an Enterprise Service Bus (ESB), such as Kong or Mulesoft, to be exposed through APIs in the Open Banking ecosystem.

Solutions: Hazelcast In-Memory Solutions

Hazelcast Cloud

The benefits of moving to the cloud are well known and applicable to virtually every industry. Hazelcast offers our customers the flexibility to deploy to the cloud on their terms, whether it's a dedicated cloud, on-premise cloud, hybrid cloud, or private cloud.

In-Memory Store

Hazelcast is a leading in-memory technology supporting Open Banking. Hazelcast is designed to provide high-availability and scalability by distributing data across multiple machines. Hazelcast enriches your application by providing the capability to quickly process, store and access data from a broad and diverse range of sources, with the speed of RAM.

An important feature in the Enterprise Edition is the High-Density Memory Store which adds the ability for Hazelcast to store very large amounts of in-memory data in each node. This scale-up feature avoids operations disruptions caused by Java garbage collection, which can severely impact the performance of your business-critical systems.

Stream Processing

Hazelcast provides distributed stream processing capabilities for building IoT and microservices-based applications. The Hazelcast architecture is high performance and low-latency driven, based on a parallel, distributed core engine enabling data-intensive applications operating at real-time speeds in support of PSD2 compliance.

Stream processing is how Hazelcast processes data on-the-fly, prior to storage, rather than batch processing, where the data set has to be stored in a database before processing. This approach is vital when the value of the information contained in the data decreases rapidly with age. The faster information is extracted from the data better.

Machine Learning and Artificial Intelligence

The speed of the Hazelcast Platform enables new levels of real-time predictive model servicing in support of delivering artificial intelligence solutions, as well as allowing real-time engineering and model retraining.

Related Resources