Learn about integrating AI into IBM Db2® for z/OS to improve operational productivity

IBM Watson® Machine Learning for z/OS® details

Flexible model development

Give data science teams the flexibility to build, train and evaluate models using their Integrated Development Environment (IDE) of choice. Or use the IBM Watson® Machine Learning for z/OS® extensive model-building features that are based on enterprise-grade open source software.

Improved productivity

Optimize data scientist productivity through extensive IBM Watson Machine Learning for z/OS model-building features. This product offers several model-building modes including notebooks, visual builders, wizards and enhanced intelligence applied to data scientist activities. Automatically normalize, handle missing values and generate data features to make even novice data scientists into experts.

Enterprise-ready AI model deployment

Operationalize predictive models within transaction applications, without significant overhead, enabling real-time insight at the point of interaction. This product offers several scoring approaches including RESTful APIs and Java and CICS integration, optimized for the highest security and performance levels on IBM Z®.

Enhanced model accuracy

Enable data scientists and engineers to schedule continuous re-evaluations of new data to monitor model accuracy over time and be alerted when performance deteriorates. Automatically refresh models to maintain model accuracy with confidence.

Production-ready machine learning

Deliver essential model versioning, auditing and monitoring as well as high availability, high performance, low latency and machine learning model automation (machine learning as a service).

Quick-start solution templates

Offer essential foundational templates for common business requirements to bootstrap your machine learning efforts. Solution templates demonstrate how machine learning can run alongside your application infrastructure to add value to key business areas including fraud detection, loan approval and IT operational analytics.

Technical details

Technical specifications

What's new?

  • Significantly improves online scoring service performance for various types of machine learning models, especially for deep learning models in an Open Neural Network Exchange (ONNX) format
  • Better integration with IBM Cloud Pak for Data
  • Enhanced simplification for installation and configuration
  • New Installation Planner that helps provide guidance for use case based installation preparation
  • IBM WMLz 2.3 Online Scoring Community Edition. This lightweight version of the WMLz scoring service provides a no-charge option enabling organizations to easily download and try the WMLz in-transaction scoring approach

Software requirements

  • z/OS 2.4, 2.3 and Db2 11 for z/OS or later
  • z/OS ICSF and z/OS OpenSSH
  • IBM 64-bit SDK for z/OS Java™ v8 SR6
  • Watson Machine Learning for z/OS IDE for Linux on Z or Linux on x86
  • Red Hat OpenShift Container Platform 4.6

Hardware requirements

  • IBM z15™, z14, IBM z13®, or IBM zEnterprise® EC12 system (1 GCP, 4 zIIP, 100 GB memory, 100 GB disk space)
  • Watson Machine Learning for z/OS IDE on Linux on Z or Linux on x86
  • 3 Master nodes (4 vCPUs, 16 GB memory, 200 GB storage in the root file system, 300 GB for image registry on one master node, 10 gbps network capacity)
  • 3 Worker node (10 vCPUs, 64 GB memory, 200 GB storage in the root file system, 10 gbps network capacity)
  • Total (6 servers, 42 vCPUs, 240 GB memory, 1.5 TB storage)

Next Steps

See how it works

Try it now