Deploy your AI models on z/OS for real-time business insights at scale.
z/OS AI Framework requires IBM z14 or higher and z/OS 3.1 base product. In addition to the IBM z/OS 3.1 base product, the deployment of AI infusion into z/OS requires AI System Services for IBM z/OS (5655-164), a new zero-charge offering that delivers support of key AI lifecycle phases including data ingestion, AI model training, inference, AI model quality monitoring, and retraining services.
AI System Services for IBM z/OS integrates seamlessly with the other AI Framework for IBM z/OS components delivered with z/OS 3.1 to build an AI platform that is designed to support initial and future intelligent z/OS management capabilities. To learn more, see
- IBM z16 — all models, including A01 (machine type 3931) and A02 (machine type 3932)
- IBM z15 — all models, including T01 (machine type 8561) and T02 (machine type 8562)
- IBM z14 — all models, including M0 n (machine type 3906) and Z
- Coupling facility (internal or external; required by EzNoSQL/VSAM record level sharing (RLS)).
- IBM z/OS 3.1 AI Framework base elements
- AI System Services for IBM z/OS
z/OS AI Framework requires any of the following IBM Z servers as well as coupling facility:
In order to install z/OS AI Framework, you will need the following software components:
For the full list of installation hardware and software requirements, see
Before you attempt to install z/OS AI Framework, make sure your system meets the hardware and software requirements. Read the following white paper for additional guidelines about these requirements.
Order and install z/OS AI Framework components with AI System Services for IBM z/OS.
AI System Services for IBM z/OS must be installed and configured on the same z/OS system where z/OSMF runs. AI System Services bundles two key components consisting of IBM Z Common Data Provider (ZCDP) and Machine Learning for z/OS Core (MLz Core). To configure the z/OS AI Framework, ZCDP must also be installed on each additional z/OS system that you intend to configure for use with the z/OS AI Framework. For ordering and packaging scenarios, see
To accelerate and optimize the setup and configuration experience, the z/OSMF-based workflows allow you to perform the configuration of the AI components from the Workflows task on the z/OSMF desktop in a common way.
The instructions for each step to configure the AI components are contained within the workflow. Click the Perform tab for each step to see the details. For an input field, you can also click the information icon next to the field to display more information about that field. You will notice that steps often have substeps. Progress through the substeps in a similar manner until each step is complete.
Once the AI components are configured in the system and the services are started, you can control the AI capabilities with the AI Control Interface for IBM z/OS (z/OS AI Control Interface), a centralized place to manage AI models for z/OS. Use the Manage tab of AI Control Interface to select the currently available AI capability that is configured for your z/OS systems. The following list represents possible actions you can take to control the AI capability.
- Display a list of the AI-powered tools (starting with Workload Manager)
- Display AI enablement status by system
- Train/retrain model by system
- Manage task specific settings
- Visualize audit data (coming soon!)
- Monitor the health status of AI framework
Read more about z/OS AI Framework.
Find information for planning and implementation of the AI-powered WLM batch initiator management offering.
Get an introduction to Machine Learning for IBM z/OS.
Get an introduction to the AI Control Interface for IBM z/OS in z/OSMF Guild Session 23.
Learn how to improve your application design by creating JSON key:value databases on z/OS.
Explore z/OSMF spotlights and key features in the IBM z/OS Management Facility ONE STOP HUB.
Deploy your AI models on z/OS for real-time business insights at scale.
Improve application design by creating JSON key:value databases on z/OS; take advantage of Parallel Sysplex technology.
A link to a related solution was added for the EzNoSQL content solution.