• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2020 Fiscal Year Final Research Report

Neuromorphic processor using superconducting adiabatic quantum flux circuits

Research Project

  • PDF
Project/Area Number 18K13801
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 21060:Electron device and electronic equipment-related
Research InstitutionYokohama National University

Principal Investigator

Ayala Christopher  横浜国立大学, 先端科学高等研究院, 特任教員(准教授) (90772195)

Project Period (FY) 2018-04-01 – 2021-03-31
Keywordssuperconductor / computing / adiabatic / EDA / Josephson junction / aqfp / neuron / tensor
Outline of Final Research Achievements

We have demonstrated the world's first adiabatic superconductor microprocessor using adiabatic quantum-flux-parametron (AQFP) logic with its data processing component operating with a clock frequency of up to 2.5 GHz. This is the fastest adiabatic circuit to date and the largest functionally meaningful AQFP circuit ever with a device count of over 20,000 Josephson junctions. Because this is superconductor technology, the circuits must be cooled to cryogenic temperatures, but even if we consider the additional overhead in cooling power, the total system power is still 80x lower than state-of-the-art commercial semiconductor technology available today. Towards the development of this microprocessor, we also established a systematic design methodology so that we can consistently create more and more complex AQFP circuits via a top-down approach and a semi-custom approach. These results set the foundation for designing more novel neuromorphic processors with various components underway.

Free Research Field

Superconductor electronics, adiabatic computing

Academic Significance and Societal Importance of the Research Achievements

我々が開発した断熱的超伝導マイクロプロセッサは、エネルギー効率と高速性の面で格段に優れている。そのため、今後世界の電力消費の大きな割合を占めるデータセンターやスーパーコンピュータなどのハイエンド機器の消費電力を大幅に低減することができる。

URL: 

Published: 2022-01-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi