IBM Z Day 2023 Focuses on the Future
- IBM Z: Building an AI-infused, cyber resilient hybrid cloud platform
- App Modernization: Accelerating application modernization and AIOps across IBM Z and cloud
- Data and AI: Leveraging existing investments with AI to improve business outcomes;
- Sustainability: Creating a sustainable infrastructure with IBM Z and LinuxONE
- IBM Z Skills Acceleration: Learn new, relevant mainframe skills
Highlights from IBM Z Day 2023
The day kicked off with “Reimagining AI for the Enterprise,” a special opening keynote featuring a discussion with Ross Mauri, general manager, IBM Z & LinuxONE at IBM, and Ian Cutress, chief analyst at More than Moore. Together, Mauri and Cutress discussed how the central characteristics of the IBM z16 mainframe, which was launched in April 2022, present unique opportunities for scalability and implementation. Mauri highlighted the growing industry interest in AI systems and how the z16 platform offers a more effective approach for businesses looking to leverage the technology in their own enterprises. “It has some very unique, very defining characteristics…that were designed from the silicon up,” Mauri explained. “One of those is an embedded AI inference accelerator that’s right on the microprocessor. We’re trying to take AI and put it as central to the core processing of the system. You can do some really amazing things with this system that’s built to scale up and scale out.” Cutress echoed how the z16 presents growth opportunities for businesses and discussed how the physical structure of the mainframe and its Telum chip generate industry-leading processing speeds, storage and security not often found on similar mainframe systems. “One of the highlights of the Telum chip is this built-in AI engine used for offloading the common matrix math we see today in modern machine learning workloads or generative AI use cases. In Telum, each core uses a dedicated interconnect to access this AI accelerator, keeping data on chip rather than relying on external system resources, and keeping latency super low,” Cutress explained. “That super low latency enables in-line use cases…that wouldn’t have been possible before.” As the session came to a close, Cutress and Mauri agreed that it’s the collaborative and curious nature of IBM users that drives forward these major mainframe breakthroughs.AI and IBM Z
AI and its implementation on individual enterprises was a focal point for many of the day’s discussions, continuing with “Revolutionizing Enterprises with AI on IBM Z: A Deep Dive into Cutting-Edge Trends and Use Cases” with IBM Fellow, CTO AI for IBM Z & LinuxONE, CTO z/OS at IBM Elpida Tzortzatos. Throughout the session, Tzortzatos explored AI’s growing popularity, its effectiveness in numerous industries and its various use cases. In particular, she highlighted IBM Z’s focus on creating a well-rounded, easily applicable system in Z that can be installed into workloads without slowing down a business’s applications. “All the technology that we delivered on IBM Z—and we continue to invest in and will deliver in the future—is all about building a very robust AI infrastructure that will allow our clients to very easily embed AI in their mission-critical workloads, while not impacting their SLAs, and doing so on the most secure, scalable and resilient platform on the planet,” she explained.
The session also delved into IBM’s investments both in proprietary AI frameworks specifically for Z as well as outside frameworks designed to pair with Z. Whether IBM Z clients train and deploy their own models through the Z platform or bring pre-trained models from another platform or framework onto Z, users can leverage IBM ONNX—the Open Neural Network Exchange—to run their models through deep learning compilers and accelerate their processes. Tzortzatos noted, “The mission of ONNX is to support model portability and allow the data scientists to have the option and flexibility to train their models in any framework and platform.”
Ultimately, Tzortzatos stressed the importance of engagement with AI technology. “Enterprises around the world are embracing AI and looking to leverage the full potential of AI to make better, more informed decisions. In my opinion, enterprises who don’t adopt and embrace AI are going to be left behind.”