Architecture Overview

Cúram is not offered as a Platform-as-a-Service (PaaS) or Software-as-a-Service (SaaS) solution in the cloud. Instead, it is a configurable platform for Social Program Management that customers tailor to meet their specific requirements. Customers are responsible for building, deploying, running, and operating Cúram in environments aligned with their own Software Development Life Cycle (SDLC) processes.

Cúram can either be deployed on traditional hosting architectures (i.e. virtual machines or bare metal) or on Kubernetes. This runbook focuses exclusively on cloud-native hosting on Kubernetes.

Cúram can be deployed on the following distributions:

*OpenShift deployments are environment-agnostic and can be run on any infrastructure that supports OpenShift, including on-premises, private cloud, or public cloud environments.

Figure 1 shows the essential nature of the Cúram architecture on Kubernetes.

It conveys the governing ideas and major building blocks of the architecture. The “Development & DevOps”, “Security”, and “Governance” components and processes described in the architecture diagram are just for reference. They will likely be different depending on your Deployment Architecture.

Cúram does not require nor impose these components in the architecture.

Figure 1: Cúram on Kubernetes

Figure 2 shows how Cúram is built as a containerized application by using WebSphere® Application Server Liberty, packaged as Docker® containers, orchestrated by Helm, and running on Red Hat OpenShift or Kubernetes Service.

Figure 2: Cúram packaging for Kubernetes

To support containerized architectures, a number of architectural changes were made. The changes are documented as follows and apply only to Cúram running on Kubernetes.

For more information about Cúram Kubernetes architecture changes, see Kubernetes architecture in the Cúram on Kubernetes Guide.