Nvidia to help develop new Japanese supercomputer

US technology company Nvidia will join the development of a new Japanese supercomputer codenamed FugakuNEXT, which is expected to be operational by 2030, according to Japanese state-run research center Riken, a co-sponsor of the project.

Nvidia will be responsible for the design of the graphics processing unit (GPU) with the aim of achieving faster computing speeds for artificial intelligence (AI), while Japan's Fujitsu, which already works with Riken, will lead the basic design of the system, the computing nodes and the central processing units (CPU), the institute detailed in a statement.

“It will be the first Japanese system to employ a GPU in the acceleration section,” said the team, which expects FugakuNEXT to offer cutting-edge computing and artificial intelligence capabilities.

The new Japanese supercomputer model is expected to be five to ten times more powerful than its predecessor, Fugaku, and accelerate application speeds "up to 100 times," capable of performing a sextillion calculations per second for the development of AI applications, according to the institute, which added that the results will be shared globally.

The new Japanese supercomputer is expected to contribute to areas such as drug development, automotive design and the search for measures against climate change .