On April 8, IBM announced the launch of IBM z17, which is the most advanced mainframe to date. It is equipped with the second-generation Telum II processor and is purpose-built for over 250 AI use cases.
IBM Prepares Mainframes for Enterprise Computing
IBM mainframes are super-powerful computers that big companies use to handle things like bank transactions, flight bookings, and healthcare records by working all day, without crashing. They are now upgraded to work with AI too.
While often viewed as legacy technology, mainframes remain indispensable. Roughly 71% of Fortune 500 companies still depend on them for critical operations, according to Market Research Future.
IBM’s z17 will reaffirm the mainframe’s relevance in the AI era by focusing more on performance gains and energy efficiency, which could change the dynamics of how companies integrate generative AI into high-volume, mission-critical systems.
The z17 can execute 450 billion AI inference operations daily, which is a 50% more than its predecessor ‘z16’. It features 48 IBM Spyre AI accelerator chips at launch, with a roadmap to double that within a year.
While z17 is designed to play well with open-source tools, existing software, and third-party hardware, the system also offers full encryption and enhanced energy savings, which uses 5.5 times less energy than comparable AI platforms, according to IBM.
Tina Tarquinio, IBM’s VP of product management for IBM Z, described the z17 as both a “culmination of long-term planning and a timely response” to the post-ChatGPT AI boom. “We’re purposely building in AI agility,” she noted, which emphasizes the system’s headroom for emerging models and memory-intensive workloads.
“The industry is quickly learning that AI will only be as valuable as the infrastructure it runs on,” said Ross Mauri, general manager of IBM Z and LinuxONE, IBM. “With z17, we’re bringing AI to the core of the enterprise with the software, processing power, and storage to make AI operational quickly. Additionally, organizations can put their vast, untapped stores of enterprise data to work with AI in a secured, cost-effective way.”
Source: https://newsroom.ibm.com/z17
Latest Stories:
