Taking a cue from China, Japan is developing a new supercomputer that could be among the world’s fastest systems when released in 2020.
The computer — being developed as part of a national project called Flagship2020 — is being developed with the aim to deliver “100 times more application performance” than the current K, which is installed in Japan and is the world’s third-fastest computer, according to the Top500 list of supercomputers, released on Monday.
The supercomputer will be deployed by 2020. It is being developed by Fujitsu and Japanese research institution RIKEN, which also developed K. The current K supercomputer has 705,204 processing cores and offers 10.5 petaflops of performance.
Details about the new supercomputer will be shared at two different sessions on Tuesday at the Supercomputing 15 conference being held in Austin, Texas.
The systems will be based on the Linux OS and the use of a “6D mesh” will be considered, according to details shared on the Supercomputing 15 website. The system will also have many storage layers, according to information on the site.
The current K is based on Fujitsu’s SPARC64 VIIIfx processors and Tofu interconnect.
The U.S., Japan and China are in a race to build the world’s fastest supercomputer. An earlier version of the K computer briefly held the title of the world’s fastest supercomputer in 2011. China’s Tianhe-2 is the world’s fastest supercomputer today, delivering peak performance of 54.9 petaflops.
Countries are rushing to develop faster computers to boast about their progress in technology, but also to boost economic, weapons and science programs.
A number of supercomputers that are faster than existing systems are on the horizon. A U.S. Department of Energy supercomputer called Aurora, due in 2019, will deliver 180 petaflops of performance. China is also planning a supercomputer of more than 100 petaflops.
The countries are hoping to create exascale supercomputers between 2018 and 2020, in which systems can conduct million trillion calculations per second.