PFN’s MN-3 Deep Learning Supercomputer Achieves Energy Efficiency of 39.38 GFlops/W, Tops Green500 for Third Time
Preferred Networks, Inc. (PFN) and Kobe University are happy to announce that, PFN’s in-house supercomputer MN-3, powered by MN-Core processor jointly developed by the research group led by Professor Jun Makino (Deputy Director of JICFuS) and PFN, topped the latest Green500 list of the world’s most energy-efficient supercomputers for the third time. The performance per watt has been further improved from the previous value.
For more detail, please refer to https://www.kobe-u.ac.jp/research_at_kobe_en/NEWS/collaborations/2021_11_16_01.html
