IBM’s next-gen mainframe chip is the first to run Arm and Z workloads on the same cores

IBM is set to revolutionize the mainframe industry with its next-generation chip, capable of seamlessly running both IBM's proprietary instruction set and Arm architecture workloads on the same cores. This groundbreaking processor, to be integrated into the upcoming IBM Z and LinuxONE systems, marks the first dual-architecture mainframe processor ever developed.

By enabling enterprises to execute the rapidly expanding ecosystem of Arm-native Linux software alongside the z/OS transaction-processing workloads that power the world's financial institutions, the chip addresses the long-standing challenge of maintaining the mainframe's relevance in the AI era.

The key engineering feat lies in the chip's design, which incorporates 11 bilingual cores capable of dynamically switching between Arm and Z software modes. This innovative approach, leveraging the open-source KVM hypervisor, allows for the coexistence of Arm64 Linux virtual machines and Linux on Z virtual machines, all running concurrently without any performance penalty. The switch between modes occurs at the nanosecond scale, effectively negating any overhead.

Built on a state-of-the-art 2-nanometer process node, the chip boasts exceptionally fast processing speeds with each core operating at a base frequency exceeding 5.7 GHz. Additionally, the chip incorporates on-chip AI inference accelerators for real-time fraud detection, a dedicated data processing unit for I/O acceleration, and a large cache architecture.

These specifications position the chip as a modern and powerful alternative to traditional mainframe technology, while still delivering the reliability and performance that businesses have come to expect.