Simulation / Modeling / Design

IBM Donates a $12 Million GPU-Accelerated Supercomputer, powered by NVIDIA V100 GPUs, to MIT

Aug 28, 2019
By Nefi Alarcon
Discuss (0)

AI-Generated Summary
  • IBM is donating a $12 million GPU-accelerated supercomputer named Satori to MIT to help researchers run more elaborate AI models.
  • The cluster comprises 256 NVIDIA V100 GPUs and 64 IBM Power 9 nodes interconnected via EFR Infiniband, and is expected to perform 2 quadrillion calculations per second.
  • Satori is modeled after the architecture of Summit, the world's fastest supercomputer, and will come online this fall alongside the opening of MIT's new College of Computing.
  • The system will be open to all members of the MIT academic community for projects such as designing next-generation hearing aids and longer-lived lithium-ion batteries.

Next Step

Powered by NVIDIA Nemotron. AI-generated content may summarize information incompletely. Verify important information. Learn more

To help MIT researchers run more elaborate AI models, IBM is donating a $12 million GPU-accelerated supercomputer, powered by the latest NVIDIA V100 GPUs, to the university. 

The new cluster named Satori, a Zen Buddhism term for “sudden enlightenment,” will come online this fall in conjunction with the opening of the university’s new College of Computing. 

The new supercomputer is modeled after the architecture of Summit, the world’s fastest supercomputer. 

Comprised of 256 NVIDIA V100 GPUs, 64 IBM Power 9 Nodes, interconnected via EFR Infiniband, Satori is expected to perform 2 quadrillion calculations per second. 

“IBM’s gift couldn’t come at a better time,” said Maria Zuber, a geophysics professor and MIT’s vice president of research. “The opening of the new college will only increase demand for computing power. Satori will go a long way in helping to ease the crunch.”

The system is specifically built for AI researchers working on projects such as designing the next hearing aid to a longer-lived lithium-ion battery.

Satori will be open to all members of the MIT academic community, the university says. 

Read more>

Discuss (0)

Tags

Simulation / Modeling / Design | Academia / Education | News | Machine Learning & Artificial Intelligence

About the Authors

Nefi Alarcon
About Nefi Alarcon
Nefi Alarcon is a senior executive communications manager on NVIDIA's leadership team. He has years of media relations and communication experience, and has previously worked at Google, Mozilla, and CNN. He received his bachelor's degree in Journalism from George Washington University.

Comments