Analyst(s): Brendan Burke
Publication Date: April 7, 2026
IBM and Arm have announced a strategic collaboration to develop dual-architecture enterprise platforms that enable Arm-based workloads to run on IBM Z systems. The move aims to expand software compatibility, improve flexibility, and support AI and data-intensive workloads in regulated, mission-critical environments.
What is Covered in This Article:
- IBM and Arm are collaborating to develop dual-architecture hardware enabling Arm-based workloads on IBM Z and LinuxONE systems
- The initiative focuses on virtualization, security, data sovereignty, and expanding software ecosystem compatibility
- The collaboration targets AI and data-intensive workloads, particularly in regulated environments where data cannot move to the cloud
- Arm’s ecosystem and efficiency advantages are being extended into enterprise mainframe environments
- The move reflects IBM’s continued investment in mainframe AI infrastructure, including Telum II and Spyre Accelerator
The News: IBM announced a strategic collaboration with Arm to develop dual-architecture hardware that allows Arm-based software environments to run on IBM Z and LinuxONE systems. The initiative is designed to support AI and data-intensive workloads by expanding software compatibility, enabling virtualization of Arm environments, and maintaining enterprise requirements such as security, reliability, and data sovereignty.
The collaboration focuses on three areas: enabling virtualization technologies for Arm workloads on IBM platforms, supporting the performance and efficiency needs of modern applications, and building shared technology layers to expand software ecosystems. IBM stated that the effort targets enterprises running regulated workloads that cannot move to the cloud, while also aiming to improve system flexibility and expand infrastructure choice without disrupting existing mission-critical environments.
IBM and Arm Partner on Dual-Architecture Computing To Redefine Mainframes for AI
Analyst Take: By partnering with IBM’s mainframe business, Arm can build on its clear intent to bring cloud price-performance to the on-premises data center. Arm’s recent AGI CPU announcement allows enterprises that use custom Arm CPUs in cloud environments to repatriate their workloads to their local data centers. A missing link in this strategy was the mainframe, a category that has experienced a resurgence due to AI. IBM’s Z mainframe product reached its highest fourth-quarter revenue in two decades in Q4 2025, contributing to $5.1 billion in infrastructure revenue, indicating that sensitive workloads are merging with AI-native processors. This partnership combines IBM’s strength in mission-critical systems with Arm’s power-efficient architecture and broad software ecosystem.
Virtualization as the primary mechanism for software expansion
The collaboration expands virtualization technologies to allow Arm-based software environments to operate within IBM enterprise platforms. This approach is intended to eliminate the need to port applications natively to IBM Z architectures, which is costly, time-consuming, and uncommon in modern development environments. By enabling Arm workloads to run via virtualization or emulation, IBM aims to expand software compatibility while simplifying how developers bring applications into mission-critical systems. However, IBM has not disclosed whether this will be implemented at the hypervisor level, through PR/SM partitioning, or via container-based approaches, leaving a key technical gap for enterprise architects. This reliance on virtualization highlights a practical pathway to software portability but also introduces unanswered questions about implementation specifics and operational trade-offs.
Extending Arm into regulated and mission-critical environments
The collaboration explicitly targets enterprises running regulated workloads that cannot be moved to the cloud, with a focus on security, data residency, and high availability. IBM’s mainframe platforms are primarily deployed in repositories of critical data, including financial systems, government databases, and high-value transactional engines. By enabling Arm workloads to run closer to these systems of record, the approach reduces latency, minimizes the need for data replication, and addresses compliance risks associated with moving data across external platforms. At the same time, Arm’s ecosystem, which already contributed 50% of server CPUs for top hyperscalers in 2025, is being extended into these enterprise environments. This positioning reflects a targeted expansion of Arm beyond cloud-native use cases into sovereign and air-gapped markets, reinforcing the role of mainframes as controlled execution environments for modern workloads.
Balancing flexibility with performance trade-offs
While the collaboration expands flexibility and software choice, it is not positioned as a performance-driven solution for all workloads. Running Arm workloads on IBM Z through virtualization or emulation introduces performance penalties, and the model is not intended for performance-intensive applications. Instead, the focus is on total cost of ownership, operational stability, and risk mitigation, which are identified as primary decision factors for enterprise customers. IBM’s hardware investments, including the Telum II processor with eight cores running at 5.5GHz and a 40% larger 360MB cache, and the Spyre Accelerator with 32 compute cores and up to 1TB of memory per IO drawer, support AI workloads at mainframe scale but do not eliminate architectural trade-offs. This highlights that the IBM Arm dual-architecture approach prioritizes integration and operational continuity over raw performance, making it suitable for specific enterprise scenarios rather than broad replacement of existing AI infrastructure strategies.
Positioning within broader enterprise infrastructure strategies
The collaboration reflects a broader effort to reposition the mainframe within modern enterprise infrastructure strategies, particularly as AI adoption increases. At first, Arm can expand mainframe use cases and make the platform more attractive to CIOs by enabling cloud repatriation without requiring code changes. Going forward, IBM is pursuing parallel AI infrastructure strategies, including its expanded collaboration with NVIDIA for GPU-based analytics and AI deployments. The alignment of AI workloads with mainframe computing places the IBM Arm dual-architecture approach as part of a broader multi-architecture strategy rather than a standalone solution.
What to Watch:
- Lack of clarity on virtualization implementation may delay enterprise adoption decisions
- No defined timeline or technical specifications for dual-architecture systems, with development potentially extending over multiple years
- Performance limitations of virtualization could restrict adoption for high-performance AI workloads
- Continued reliance on GPU-based infrastructure, including IBM’s NVIDIA collaboration, may limit the role of Arm in large-scale AI deployments
- Enterprise adoption will depend on the ability to integrate Arm workloads without disrupting existing mainframe operations
See the complete press release about the IBM and Arm strategic collaboration to develop dual-architecture enterprise platforms on the IBM website.
Declaration of generative AI and AI-assisted technologies in the writing process: This content has been generated with the support of artificial intelligence technologies. Due to the fast pace of content creation and the continuous evolution of data and information, The Futurum Group and its analysts strive to ensure the accuracy and factual integrity of the information presented. However, the opinions and interpretations expressed in this content reflect those of the individual author/analyst. The Futurum Group makes no guarantees regarding the completeness, accuracy, or reliability of any information contained herein. Readers are encouraged to verify facts independently and consult relevant sources for further clarification.
Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.
Other Insights from Futurum:
Arm’s $15 Billion CPU Opportunity Hinges on Agentic Data Center Design
Arm Q3 FY 2026 Earnings Highlight AI-Driven Royalty Momentum
IBM Q4 FY 2025: Software and Z Cycle Lift Growth and FCF
Image Credit: Arm
Author Information
Brendan is Research Director, Semiconductors, Supply Chain, and Emerging Tech. He advises clients on strategic initiatives and leads the Futurum Semiconductors Practice. He is an experienced tech industry analyst who has guided tech leaders in identifying market opportunities spanning edge processors, generative AI applications, and hyperscale data centers.
Before joining Futurum, Brendan consulted with global AI leaders and served as a Senior Analyst in Emerging Technology Research at PitchBook. At PitchBook, he developed market intelligence tools for AI, highlighted by one of the industry’s most comprehensive AI semiconductor market landscapes encompassing both public and private companies. He has advised Fortune 100 tech giants, growth-stage innovators, global investors, and leading market research firms. Before PitchBook, he led research teams in tech investment banking and market research.
Brendan is based in Seattle, Washington. He has a Bachelor of Arts Degree from Amherst College.
