Americas

  • United States
Jon Gold
Senior Writer

IBM brings AI to Z series mainframes, z/OS, Cloud Paks

News
Sep 26, 20232 mins
Artificial IntelligenceServers

A new range of AI integrations is coming to IBM’s mainframes, operating systems and private clouds.

IBM Z mainframe
Credit: Connie Zhou for IBM

New AI toolkits, machine learning (ML) frameworks and AI-based private cloud tools are on their way to IBM Z-series mainframe users, as the company looks to preserve its share of the fast-growing AI marketplace.

The company announced today that its newest offerings, meant to help organizations get to work on the latest and greatest in AI frameworks, will be available for IBM Z, LinuxOne, z/OS and Cloud Pak architectures. The first new feature, AI Toolkit for IBM Z and LinuxONE, is designed to help enable mission-critical business applications on open-source frameworks like IBM’s Z Accelerated line with TensorFlow, SnapML and more. A Python AI Toolkit for the underlying z/OS also provides an open-source Python library for AI/ML workloads, designed to conform with the company’s Security and Privacy by Design guidelines.

“These tools include frameworks and libraries that are optimized and supported for IBM Z and they are built to allow developers to start implementing trustworthy AI capabilities on z/OS,” IBM said in a statement. “Supported by the same underpinnings for watsonx.ai that empower data scientists and developers to build, run, and manage machine learning models, these toolkits will evolve over time to become an integral part of the watsonx platform.”

New versions of IBM’s z/OS Enterprise and Core Editions will also be available, meant for data scientists and developers looking to create ML and deep learning models for their operational data. The platform, according to IBM, is effectively a watsonx.ai extension, bringing the company’s generative AI capabilities to z/OS. A new Auto AI tool will also be released for IBM’s Cloud Pak private cloud software.

“This allows users to upload their data, choose the problem type and specify constraints, and run a series of automated experiments that generate a range of high performing pipelines quickly and easily,” the company’s statement said.

Finally, operating system-level AI tuning is coming to z/OS 3.1’s software layer, providing automated learning and predictions for the optimization, management and performance enhancement of IT processes.

The new ML features for z/OS, Auto AI for Cloud Paks, and Python libraries are available today, according to IBM. The AI-infused z/OS 3.1 features will be released on September 29, while the AI toolkit for IBM Z and LinuxOne will be released sometime in the fourth quarter. The company declined to provide pricing information.