2 About Prerequisite Software Requirements

Learn about the required operating system, software, tools, and files required to deploy artificial intelligence (AI) and machine learning (ML) services in Oracle Communications Cloud Scale Monetization.

Topics in this document:

System Requirements

AI and ML services in Oracle Monetization Suite require a Linux operating system. To ensure efficient deployment and operation, use the current, supported versions of all required software, tools, platforms, and reference files.

For details and a list of supported versions, see BRM Compatibility Matrix.

Setting Up Prerequisite Software and Tools

Before you deploy data, training, and prediction services, you must install and configure the following software and tools:

  1. Kubernetes Cluster: Set up a Kubernetes cluster to orchestrate and manage your containerized service deployments.

    For more information about Kubernetes, see "Kubernetes Concepts" in the Kubernetes documentation.

  2. Container Runtime: Install a Kubernetes-supported container runtime such as Podman on all cluster nodes to support containerized workloads.

  3. Helm 3: Install Helm 3 to deploy and manage services using Helm charts.

    For installation instructions, see "Installing Helm" in the Helm documentation.

  4. Kubernetes Namespaces: Create the following namespaces for separation of services:

    • data-fetch-service for the data service

    • training-utility-service for the training service

    • predict-service for the prediction service

  5. Oracle Cloud Infrastructure (OCI) Policies: If you plan to deploy services in OCI, create and assign necessary OCI policies to enable access between Object Storage, Logging, Data Flow, and Data Science.

    A policy defines who has access to specific OCI resources and what actions they can perform. For more information, see "Getting Started with Policies" in Oracle Cloud Infrastructure Documentation.

    Note:

    This step is required only for deployments in OCI.

  6. Oracle Identity Cloud Service (IDCS): Obtain access to IDCS, which serves as an identity provider for user authentication and access management.

    For more information, see "Federating with Identity Providers" in the OCI documentation.

  7. Prometheus: Install Prometheus to monitor service metrics and trigger alerts.

  8. Grafana: Install Grafana for visualizing metrics, importing dashboards, and analyzing data.

  9. Tempo: Install Tempo for distributed tracing to monitor and troubleshoot services.

  10. (Optional) NGINX Ingress Controller: Install the NGINX Ingress Controller.

    For installation instructions, see Installation Guide in the Kubernetes documentation.

    Note:

    If you are not using the NGINX Ingress Controller, configure direct HTTPS with an SSL certificate and private key. This is the recommended approach.