Why IBM and Arm are partnering up on AI for mainframes

IBM and Arm are teaming up to adapt Big Blue's enterprise hardware to run Arm-based software, as companies look to bridge the gap between legacy mainframe infrastructure and the AI era.

The two companies are joining forces to develop dual-architecture hardware for IBM’s mainframe business. The first product they are focussing on is virtualisations to run AI software, based on Arm's chip architecture, on IBM's on-prem infrastructure.

Eddie Ramirez, Arm Cloud AI Business Unit’s VP of GTM, told The Stack the partnership would extend Arm’s ecosystem, currently not compatible with IBM’s mainframes, into “new areas of enterprise computing.”

Moor Insights Analyst Patrick Moorhead said AI is reshaping what enterprises need from infrastructure: "Flexibility, workload portability, and ecosystem reach are becoming just as critical as performance and reliability."

Arm and IBM's partnership points toward a future where platforms "can evolve without forcing disruptive tradeoffs," he added.

Bringing AI to traditional systems

Ramirez said, “There is growing interest across the industry in bringing AI capabilities closer to where enterprise data resides, including within traditional systems like mainframes.