Production

Minimal latency for all processing steps is crucial for production of Immediate Payments applications performance.

The application must be available to its users any time, so the high availability (HA) concept to be put in place can allow any major downtime of this payment instrument. With the different supported platforms and high availability options, IBM® offers all levels of systems for hosting this mission-critical application.

Depending on the detailed HA requirements, any systems must be intensively tested before they are moved to production. The physical availability of adequate test systems and the full coverage of exception scenarios within the test period are important for verifying the performance and HA characteristics of any production environment.

In production, the nature of an immediate payment should be seen as a combined all or nothing requirement. If any major problem in the process chain occurs, the payment is most likely not going to be successful because of the timeout conditions that apply for Immediate Payments transactions. From a system point of view, in-flight payments might be canceled or are allowed to time out if technical problems occur. This leads the system to return immediately to a stable production environment without being too busy to recover payments that were going to be invalid.

Typical areas that need to be considered for the best performance and HA characteristics are:
  • Definition of enough hardware resources for running the peak workloads. Because of critical latency requirements, systems should not exceed 80% of their targeted resource consumption.
  • Introduction of data sharing and data replication capabilities. This typically enables the application to be spread over multiple sides and offers a high extend of fault tolerance.
  • FTM is based on IBM App Connect Enterprise, which offers various options for high availability configurations on different hardware platforms. This allows the definition of system characteristics according to specific preferences and needs.
  • The planning of particular high availability configurations defines whether load balanced, hot standby, or cold standby configurations are required by the specific business needs. This drives the complexity and cost of the production environments, but also defines the level of availability that is achievable with the selected configuration option.