×
Emperor’s New Groove: IBM unveils AI-powered LinuxONE Emperor 5 mainframe
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

IBM‘s new LinuxONE Emperor 5 mainframe represents a significant fusion of traditional enterprise computing with modern AI capabilities, signaling a strategic shift in how massive computing resources can be deployed for mission-critical AI workloads. This fifth-generation system introduces unprecedented levels of processing power, security features, and cost efficiency while addressing the growing enterprise demand for dedicated, high-performance AI infrastructure that doesn’t sacrifice the reliability mainframes are known for.

The big picture: IBM’s LinuxONE Emperor 5 transforms the concept of mainframes from legacy technology into a cutting-edge AI powerhouse tailored for enterprise-grade computing needs.

  • The system features the new IBM Telum II processor, manufactured using Samsung‘s 5nm process technology, delivering substantial performance improvements with eight cores running at 5.5GHz.
  • This represents IBM’s strategic bet that enterprise AI adoption requires purpose-built secure infrastructure rather than just cloud-based services.

Key technical specifications: The Telum II processor features 40% more on-chip cache capacity and a dedicated AI accelerator delivering four times the computing power of previous iterations.

  • The integrated AI accelerator can perform up to 24 trillion operations per second (TOPS), specifically designed for enterprise AI workloads.
  • Virtual L3 and L4 caches have been expanded to 360MB and 2.88GB respectively, significantly improving data processing capabilities.

Advanced AI capabilities: Beyond the built-in processing power, IBM is expanding the system’s AI potential with additional specialized hardware and software.

  • The forthcoming IBM Spyre Accelerator, expected in late 2025, will add 32 AI accelerator cores that can be connected via PCIe, allowing for modular AI expansion.
  • An updated AI Toolkit optimized for the Telum II processor aims to streamline developer productivity and AI deployment in enterprise environments.

Security enhancements: The LinuxONE Emperor 5 incorporates extensive security features tailored for sensitive enterprise AI applications.

  • The system includes confidential computing capabilities that keep data encrypted even while in memory, alongside NIST-standardized post-quantum cryptographic algorithms.
  • Additional security features include a hardware security module, support for confidential containers, and integration with IBM Vault Self-Managed for secrets management.

Business impact: IBM positions the system as delivering significant operational efficiencies compared to traditional x86 server deployments.

  • The company claims the LinuxONE 5 can reduce total cost of ownership by up to 44% over five years compared to x86 alternatives through workload consolidation.
  • The system offers 99.999999% availability (eight nines), addressing critical requirements for business continuity in AI-dependent operations.

Software ecosystem: The mainframe will run specialized Linux distributions optimized for the platform’s unique architecture.

  • Red Hat Enterprise Linux (RHEL) for LinuxONE serves as the primary operating system for the platform.
  • Technology previews of Red Hat OpenShift AI and OpenShift Virtualization are now available, enabling organizations to manage both traditional VMs and containerized workloads through a unified interface.
IBM introduces a mainframe for AI: The LinuxONE Emperor 5

Recent News

NYT strikes landmark AI licensing deal with Amazon

The prestigious newspaper establishes a template for how media organizations might monetize content in the AI era while still pursuing litigation against other technology companies.

AI chip startup Cerebras outperforms NVIDIA’s Blackwell in Llama 4 test

Cerebras's custom AI hardware delivers more than double the tokens per second of NVIDIA's Blackwell GPUs in independent testing of Meta's largest language model.

AI courses from Google, Microsoft and more boost skills and résumés for free

As AI becomes critical to business decision-making, professionals can enhance their marketability with free courses teaching essential concepts and applications without requiring technical backgrounds.