IBM has announced the z17, its next-generation mainframe designed for the age of artificial intelligence (AI). The system is powered by the new IBM Telum II processor, enabling AI capabilities across hardware, software, and system operations. This next-gen mainframe is engineered to support enterprise workloads with AI acceleration, making it easier for organizations to scale their AI-driven tasks, according to a statement.
The IBM z17 is poised to redefine AI at scale, offering the ability to process up to 50% more AI inference operations per day than its predecessor, the z16. Businesses will benefit from the system’s ability to run over 250 AI use cases, including applications in industries such as financial services, healthcare, retail, and more. For example, it can be used to mitigate loan risks, manage chatbot services, support medical image analysis, and prevent retail crime.
The z17’s development represents over five years of design and innovation, including the filing of more than 300 patent applications with the U.S. Patent and Trademark Office. More than 100 clients provided input on the system, working closely with IBM’s Research and Software teams to ensure it meets enterprise needs. The result is a mainframe with enhanced capabilities, such as multi-model AI processing, new security features, and tools that improve system management and usability.
One of the key features of the z17 is its integration of AI inferencing capabilities, powered by the Telum II processor’s second-generation on-chip AI accelerator. This allows the system to process 450 billion AI inferences daily with a response time of just one millisecond. The z17 also includes the IBM Spyre Accelerator, set to be available in Q4 2025, which will further enhance AI compute capabilities, particularly for generative AI applications such as virtual assistants running on enterprise data.
IBM has also focused on enhancing the user experience with AI-driven tools. The z17 introduces IBM watsonx Code Assistant for Z and IBM watsonx Assistant for Z, which help developers and IT operations teams boost productivity and streamline system management. For the first time, watsonx Assistant for Z will be integrated with Z Operations Unite, providing AI chat-based incident detection and resolution using live systems data.
Security remains a top priority, with IBM z17 introducing several new AI-powered features. These include advanced data protection capabilities, powered by AI-driven detection and classification of sensitive information, and improved secrets management via IBM Vault. IBM Vault helps manage the entire lifecycle of sensitive data, providing a single solution to protect critical workloads across hybrid cloud environments.
The z17 also integrates IBM’s latest storage solution, IBM Storage DS8000, which is optimized to work seamlessly with the system, providing secure, agile, and scalable storage for mission-critical workloads. This integrated infrastructure delivers high-performance data management and enhances the security of enterprise operations.
The IBM z17 is set to be generally available on June 18, 2025. The IBM Spyre Accelerator is expected to be available starting in Q4 2025.
Ross Mauri, General Manager of IBM Z and LinuxONE, said, “The industry is quickly learning that AI will only be as valuable as the infrastructure it runs on. With z17, we’re bringing AI to the core of the enterprise, providing the processing power and security to make AI operational quickly and securely.”
With the z17, IBM aims to transform the enterprise AI experience, enabling businesses to unlock the full potential of their data while maintaining the reliability, security, and scalability that IBM Z systems are renowned for.
Bd-pratidin English/ Jisan
