The Rensselaer Polytechnic Institute in New York is home to IBM Corp.’s latest new supercomputer.

The research university said today that it plans to use the supercomputer to accelerate its research and development of new artificial intelligence computing systems.

IBM’s latest eight-petaflop supercomputer is called AiMOS, which stands for Artificial Intelligence Multiprocessing Optimized System. It can run up to 8 quadrillion calculations per second. That makes AiMOS the most powerful supercomputer ever installed in a private university.

It also ranks 24th in the world in the Top 500 twice-yearly list of the world’s most powerful nondistributed computer systems. Interestingly, though, it runs exactly the same Power Systems architecture as the world’s current two most powerful supercomputers, Summit and Sierra, which are operated by the U.S. Department of Energy.

The university said it’s planning to use AiMOS to help develop new hardware that can be used to support AI applications.

“Our collective goal is to make AI systems 1,000 times more efficient within the next decade,” said IBM Executive Vice President John Kelly.

As AI algorithms become increasingly more complex, the hardware platforms that run them also need to be much more powerful to run them effectively. That means making advancements in critical technologies such as central processing units, graphics processing units and accelerator chips, and that’s just what AiMOS is going to help with.

Researchers at Rensselaer intend to use the supercomputer to aid in the design of new computer chips and systems. It will provide the “modeling, simulation and computation necessary to support the development of this hardware,” IBM said.

“In order to realize AI’s full potential, special-purpose computing hardware is emerging as the next big opportunity,” Kelly added.

In addition, AiMOS will be used to support existing projects at Rensselaer. They include the Jefferson Project, an environmental-monitoring initiative that sucks up more than 9 terabytes of data annually from sensors in place at Lake George in New York.

Photo: Dan Farber/Flickr