Powerful supercomputer to advance AI research in UK

Isambard-AI “a ground-breaking moment for science, innovation and technology,” Viscount Camrose

Announced as the UK’S fastest and most powerful supercomputer, the Isambard-AI at the University of Bristol is now officially online and accessible to a wide range of UK businesses. Having been placed in the global power rankings for the fastest, it has moved into the top two of the ISC's Green500 list and also appeared at position 129 in the TOP500 list of high-performance computing (HPC) systems on Earth, as published at the ISC High Performance 2024 event in Hamburg, Germany.

With the latest technologies, including HPE Slingshot 11 interconnect and direct liquid cooling solutions with highly-integrated, heterogeneous CPU-GPU systems from NVIDIA, it is also said to be one of the most efficient supercomputers that has ever been built.

Supporting UK organisations

The new Isambard-AI facility will be available to wide range of organisations to utilise the power of AI, which is already propelling emerging technologies such as training large language models (LLMs), healthcare and robotics and is expected to essentially support critical areas including automated drug discovery and climate research.

Professor Simon McIntosh-Smith, Director of the Bristol Centre for Supercomputing at the University of Bristol, explained: “Assuming there are eight billion people on earth, and everyone performed one calculation per second, it would take 2.3 years for all eight billion people, working 24/7, 365-days a year, to do what Isambard-AI phase one could do in one second. “That’s a pretty astounding performance, even though we can pack it into a relatively small space.

“Isambard-AI phase 1 signifies the start of the Isambard-AI service. When the remaining 5,280 GPUs arrive at the University’s National Composites Centre (NCC) later in the summer, it will increase the performance by a factor of 32.”

With a £225m investment from the UK Government last autumn and with thanks to HPE modular data centre technology, the Isambard-AI phase one with HPE Cray EX2500 supercomputer and 168 NVIDIA GH200 Grace Hopper Superchips was delivered in record time.

Advancing research possibilities

Minister for AI, The Viscount Camrose said: “With the launch of the first stage of University of Bristol’s supercomputer Isambard-AI, we're witnessing a groundbreaking moment for UK science, innovation, and technology. This world-class equipment will revolutionise research possibilities here in the UK, with our world-first AI Safety Institute among the organisations who are set to benefit by harnessing one of the most powerful computer systems on the planet to drive forward their vital research.

“Not only does Isambard-AI rank among the world's fastest supercomputers, but it also sets the standard for eco-conscious computing, leading the charge in efficiency and sustainability. From AI safety to healthcare and climate research, its capabilities are unparalleled, marking the UK out as a global leader in AI and responsible innovation.”

“It is exciting to see Isambard-AI entering the first stage of its journey to deliver Europe's largest AI system for open science and propel the UK as a global leader in AI," said Matt Harris, managing director for UK, Ireland, Middle East and Africa at HPE. "With AI progressing at rapid pace, we are proud to have delivered the system in a record time, with just 3 months between concept design and the system going live. This unique supercomputer is the centrepiece of the UK Government's AI Research Resource and will enable organisations like the AI Safety Institute to train generative AI models at scale with research outcomes expected as soon as May this year."

Rod Evans, EMEA Vice President of HPC and AI at NVIDIA added: “AI is rapidly advancing and driving breakthroughs across industries, from healthcare to climate research. The University of Bristol’s Isambard-AI system, powered by 168 NVIDIA GH200 Grace Hopper Superchips, will provide UK researchers with the massive AI computing capabilities needed to push the boundaries of scientific discovery.”

More information online