IBM (IBM.US) released the world's first dual-architecture mainframe processor with a single core compatible with both IBM and ARM (ARM.US) instruction sets

The Zhitong Finance App learned that at the annual Hot Chips conference held on Monday, IBM (IBM.US) unveiled the world's first dual-architecture mainframe processor. The chip enables enterprises to run applications simultaneously on both IBM and ARM (ARM.US) computing platforms to truly achieve seamless collaboration across architectures.

This new processor allows enterprises to run z/OS and Linux systems under the IBM Z platform while also running Linux systems with native Arm architecture, thus taking advantage of IBM's traditional advantages in terms of security, reliability, and scalability.

The chip uses a 2 nm process and integrates 11 high-performance cores. IBM said its main frequency can exceed 5.7 GHz, and it has a built-in AI inference accelerator for fraud detection during transactions, a dedicated on-chip data processing unit for I/O acceleration, and a large cache architecture for high-load enterprise applications. IBM further stated that each core can simultaneously execute instructions from the ARM and IBM architectures.

“As companies continue to modernize their application portfolios and integrate AI into core business processes, they need infrastructure that provides more choices,” said Christian Jacobi, IBM executive. This processor marks a major architectural breakthrough for IBM Z and LinuxONE. By bringing native Arm support to our platform, we are not only providing users with access to one of the fastest growing software ecosystems in the industry, but also continuing the core quality of IBM systems as the cornerstone of today's business operations.”

Arm executive Mohamed Awad stated, “As AI applications continue to expand, the computing field is increasingly converging on Arm architectures. IBM Z and LinuxONE support some of the most challenging workloads in highly regulated industries around the world. Bringing Arm computing power and its software ecosystem to these platforms will further push this trend into mission-critical enterprise infrastructure and provide enterprises with more choices in how to deploy AI.”