IBM launches the IBM z17, its first AI-focused mainframe.

IBM has taken a decisive step toward the full integration of artificial intelligence into its mainframe systems with the introduction of the IBM z17 , the next generation of its renowned platform. This advancement represents a milestone for all types of sectors that rely on real-time data processing with a high level of security and reliability.

This new mainframe is designed from the ground up to enable seamless interoperability between hardware , software, and AI-powered operations . This will allow companies to incorporate automated processes and predictive analytics directly onto the platform without moving sensitive data to other environments, thus enhancing security and operational efficiency. Furthermore, advancements in computer technology are crucial to understanding their impact on artificial intelligence.

An architecture built entirely for AI

The IBM z17 comes equipped with the IBM Telum II processor , a significant evolution that incorporates AI accelerators directly integrated into the chip. This design allows it to execute more than 450.000 billion AI inference operations every day, with a latency of just one millisecond. This performance is estimated to represent a 50% improvement over its predecessor, the z16. This improvement is fundamental in the context of high-performance computing.

The idea of ​​bringing AI closer to the data, rather than the other way around, is one of the key strategic principles behind the design of the z17 . This avoids the need to transfer sensitive data outside the system for analysis, which has positive implications in terms of privacy, security, and regulatory compliance.

Developed over five years, this mainframe is the result of a collaboration between more than 100 customers and various IBM research and software teams . During that time, more than 300 patent applications were filed related to its processing and AI capabilities, reflecting the volume of innovation behind it.

Another element that expands the capabilities of the z17 is the IBM Spyre Accelerator , an additional component in PCIe card format planned for the last quarter of 2025. Its objective is to offer support for multimodal AI models, including generative functions and operational assistants integrated directly into the platform.

The combination of Telum II and Spyre Accelerator will allow organizations to work with optimized environments that leverage the raw power of the hardware while integrating AI software without compromising the security of their critical data. As computing evolves, the goal is to achieve greater levels of efficiency.

An operating environment focused on intelligent automation

The z17 not only stands out for its technical architecture, but also enhances the user experience thanks to its IT environment automation capabilities . Tools like IBM watsonx Code Assistant for Z and watsonx Assistant for Z allow development and operations teams to use powerful assistants that facilitate programming , issue resolution, and intelligent chat interaction.

For the first time, these assistants will be integrated with IBM Z Operations Unite , a platform designed to gather operational data from multiple sources in OpenTelemetry format. This data collection enables real-time anomaly detection and automated action, improving response times and reducing the effort required for system management.

In addition, z/OS 3.2 , the next version of the IBM Z operating system, has been announced and is slated for release in the third quarter of 2025. This update expands support for modern tools, including NoSQL databases and hybrid data access, further strengthening the platform's ability to connect with diverse environments.

This new operating system has been optimized to maximize the AI-based capabilities of the hardware , offering richer operational information and predictive functionalities to improve real-time decision-making . The integration of artificial intelligence into operating systems is a step forward in the evolution of computing.

With all these elements, IBM has sought to make the z17 not just a version upgrade , but a true transformation of the mainframe systems operating model that combines automation, artificial intelligence and compatibility with current hybrid architectures.

Enhanced security and protection of sensitive data

One of the fundamental pillars of the IBM z17 is its focus on resilience and cybersecurity. The platform integrates new features designed to protect data in complex environments , such as its partnership with HashiCorp to incorporate IBM Vault, a solution for managing secrets in hybrid clouds.

This functionality enables authentication and authorization of access to sensitive data, keys, tokens, and certificates from a single solution integrated directly into the infrastructure, facilitating standardized secrets management that protects critical workloads.

Identifying sensitive data is also a key aspect of the system . Thanks to the use of natural language processing and intelligent algorithms, the z17 can discover, classify, and protect high-value business information without manual intervention—crucial given the rise in threats and data breaches.

Similarly, the IBM Threat Detection for z/OS solution offers an advanced defense framework that detects anomalous behavior within the system, which is especially useful for identifying potential cyberattacks before they materialize.

All of these systems are designed to work together, allowing the z17 to act as a platform capable of anticipating risks and protecting business activity without disrupting operations.

Prepared for hybrid environments and future needs

The z17 was designed to integrate seamlessly into hybrid infrastructures , recognizing that most businesses today operate in environments where on-premises and cloud systems coexist. Therefore, IBM has opted for open standards and compatible tools to facilitate a smooth implementation across diverse technology architectures.

In addition to its central processor and accelerator card, the IBM z17 works in conjunction with the latest generation of the DS8000 storage system , a solution specifically developed to complement the mainframe's performance and facilitate the management of large volumes of data.

This combination ensures consistent performance even in scenarios where workloads are particularly demanding, making it ideal for industries such as finance, healthcare, telecommunications or retail , where fast, secure and accurate data analysis is critical.

On the other hand, developers using the z17 environment will also benefit from AI-powered automation tools, which not only optimize human resources but also accelerate the implementation of new functionalities in their business systems.

The new IBM z17 is scheduled to be available on June 18, 2025 , and many companies have already begun planning its incorporation as part of their digital transformation strategies focused on artificial intelligence.

Thus, the IBM z17 is positioned as a comprehensive offering that combines processing power, intelligent automation, secure infrastructure, and adaptability to the new requirements of the AI-driven digital economy. An evolution of the classic mainframe that, far from becoming obsolete, is reinventing itself to take on a key role in the technological future of businesses.