NASA Wants To Go Into Space With RISC-V

Abdul Ghani

NASA's future space flight computers will use RISC-V cores. These should be 100 times faster than previous computers. The RISC-V specialists from SiFive will in the future supply the US space organization NASA with their own computing cores, as the CPU designer has announced. These cores are intended as the main component of NASA's high-performance spaceflight computing (HPSC) processor. NASA only announced the HPSC project in July of this year. The aim is to create new computers for the space flights themselves, i.e. for use in probes or robot vehicles and the like.
Rockets of the Future: Ares I, Ares I-X (NASA, Ares I-X, 5/28/09).NASA's Marshall Space Flight Center From Flickr

NASA writes that electronic devices in space are often damaged by radiation, which can lead to errors. In addition, depending on the location, radio signals to the earth often require a very long time, so the computers on-site would have to be able to run without being controlled by the earth. The HPSC project wants to solve these problems and also use particularly energy-efficient technology.

According to the announcement, the HPSC will now be based on the SiFive Intelligence x280 platform. Eight special RISC-V cores for vector instructions are to be used as well as four additional RISC-V cores. This should increase the theoretically possible computing power to a hundred times what can be achieved with today's space computers.

This increase is probably only possible by adapting the software to the vector units and using machine learning models adapted to them. SiFive himself speaks of an optimized version of Tensorflow Lite for the x280 platform as well as numerous models that have already been ported. In addition to NASA, the HPSC and x280 platform will also be available to other government agencies in the future.

Comments / 2

Published by

Member Of Freelancers Union (USA), Freelance Writer!, and Digital Creator. Ghani Mengal is an enthusiast Freelance blogger and digital marketer. His content has been published and featured on many popular blogs, websites, and publications. Including TeelFeed,, Data-Driven Investor, TextSniper, Scientific Publication The Predict, The Startup, The Ascent, Heart Affairs, Illumination, And The List goes on.


More from Abdul Ghani

Comments / 0