IBM releases the latest version of its mainframe hardware, which contains new updates that are supposed to accelerate the AI acceptance.
The hardware and consulting firm announced on Monday IBM Z17, the latest version of its mainframe computer hardware. This fully encrypted mainframe is driven by an IBM Telum II processor and designed for more than 250 AI applications, including AI agents and generative AI.
Mainframes seem to be like old hat, but today they are used by 71% of the Fortune 500 companies. After a source. In 2024, the mainframe market was worth 5.3 billion US dollars per future market research.
The Z17 can process 450 billion inference operations in one day, an increase of 50% compared to its predecessor. The IBM Z16which was published in 2022 and ran on the company’s original Tellum processor. The system is designed in such a way that it is able to integrate completely into other hardware, software and open source tools.
Tina Tarquinio, VP of product management and design for IBM Z, Techcrunch said that this mainframe -upgrade has been in the works for five years Opics AAIP In November 2022.
IBM spent more than 2,000 research hours to maintain feedback from over 100 customers in building the Z17, Tarquinio said. She finds it interesting to see that five years later, the feedback that they enabled them with the path agreed on the way.
“It was wild to know that we are introducing a AI accelerator and then, especially in the later half of 2022, all changes in the industry in relation to AI,” Tarquinio told Techcrunch. “It was really exciting. I think the biggest point was (that) We do not know what we don’t know, what is coming, right? So the possibilities are really unlimited in relation to what ai can help us.”
The Z17 is set up to adapt and take into account where the AI market leads, said Tarquinio. The mainframe supports 48 IBM Spyre Ai Accelerator Chips in the publication, with the plan to bring this number to 96 within 12 months.
“We intentionally build in the head,” said Tarquinio. “We deliberately build up in an agility. If new models are introduced, we make sure that we have built in headroom for larger, larger models – models that may need more local memory to speak.
Tarquinio said that one of the highlights of these latest hardware objects joked that it was as if they were asked to choose their favorite child-that the Z17 is more energy-efficient than its predecessor and supposedly also competitors.
“On-chip, we increase AI acceleration by seven and a half times, but that is five and a half less energy than you would have to do, like a multi-model in a different kind of accelerator or platform in the industry,” said Tarquinio.
The Z17 mainframes will be generally available on June 8th.
