Summit and Sierra are the world's most powerful supercomputers, and now Pangea III joins their ranks as the world's most powerful commercial supercomputer. Global business leaders put these systems to work tackling the world's biggest challenges while assessing the opportunity for increased revenue—all using technologies available to help your business tackle your own challenges.

The history of Supercomputing at IBM

IBM has a rich history of supercomputing dating back to the dawn of the computer edge, including such highlights as the Apollo program, Deep Blue, Watson, and more.


The Summit AI supercomputer by the numbers

IBM is the only vendor that has multiple systems on the list of Top 10 supercomputers, including the overall number one and two most powerful supercomputers in the world (Summit and Sierra, respectively).

200

quadrillion calculations per second

9216

IBM POWER9 CPUs

25

Gigabytes per second between nodes

27648

NVIDIA Tesla GPUs

What are supercomputers?

Supercomputers are the physical embodiment of high-performance computing (HPC), allowing organizations to solve problems that would be impossible with regular computers and to better assess associated revenue streams. Pangea III was built to show that these super computers have commercial applications, and aren't just for the large government or research organizations.

Today, supercomputing is more and more synonymous with AI. Summit and Sierra were all built with AI workloads in mind, as harbingers of what IBM calls “Chapter Two” of digital transformation at scale for enterprise AI and hybrid cloud.

Supercomputers at work

How can all of this computing power be put to work? Here are some examples.

Combatting Cancer

Combatting Cancer

Machine learning algorithms scaled on Summit will help supply medical researchers with a comprehensive view of the U.S. cancer population at an extreme level of detail.

Identifying next-generation materials

Identifying next-generation materials

Deep learning could help scientists identify materials for better batteries, more resilient building materials and more efficient semiconductors.

Accelerating understanding of disease

Accelerating understanding of disease

Using a mix of AI techniques, researchers will be able to identify patterns in the function, cooperation, and evolution of human proteins and cellular systems.


Join IBM at SC19 in Denver, Colorado

Choose from a wide range of single client briefings, engaging user groups, and demonstrations of the latest technologies driving the HPC transformation.


POWER9: The secret behind supercomputing

AC922 is the processor-based accelerated computing server that gives Summit and Sierra their strength and speed, running on POWER9™ processors. With AC922, your enterprise can confidently employ data-intensive workloads, like deep-learning frameworks and accelerated databases, thanks to an IT infrastructure designed for enterprise AI.


IBM Spectrum LSF Suites—workload management for supercomputing

IBM Spectrum LSF Suites is the comprehensive workload management solution that supports Summit and Sierra. With IBM Spectrum LSF Suites you’ll be able to simplify your own HPC through policy driven scheduling and central management of large, distributed systems.


Meet the world’s most powerful commercial supercomputer

Pangea III, built by IBM for Total, is the world’s most powerful commercial supercomputer, allowing the global energy company to more accurately locate new resources and better assess the potential associated revenue opportunities. Learn more about how IBM supercomputing is changing computing . . . and changing the world.

Resources

The story of Summit

Learn more about how IBM Power Systems and OpenPOWER experts worked with Oak Ridge National Laboratory to build Summit.

AI supercomputing

Watch an interview with IBM Exascale lead Dave Turke, and learn about HPC infrastructure and the future of AI workloads.

Supercomputer vs mainframe

Mainframe computers are less powerful than supercomputers, but they are also more affordable.