High-performance computing technologies based on superconducting circuits are actively developing due to their high energy efficiency and performance. In this article, recent research developments on low-power, high-performance superconducting integrated circuit technologies are reviewed.
Recently, several new ultra-low-power superconducting circuits, which can contribute to drastic improvement of power efficiency of information processing, have been proposed and implemented. These power-efficient superconducting circuits can be divided into dc-biased and ac-biased circuits. This paper overviews the current status of the development of power-efficient superconducting digital circuit technologies mentioning our researches mainly. Moreover, hybridization of dc-biased and ac-biased superconducting circuits is discussed for building low-power and high-performance digital information processing systems.
Summary form only given. Superconducting computing has recently emerged as a potential solution for energy-efficient, large-scale computing. Problems in large-scale computing are reviewed. Superconducting computing is significantly different from conventional technologies and required a feasibility study [1] to evaluate its potential. Concepts were investigated for computing systems capable of performance in the range from 1 to 1000 PFLOP/s. The concept systems were constrained to use existing commercial cryogenic refrigerators and Nb superconducting technology. Superconducting computing was shown to be potentially competitive on the basis of power and energy efficiency if key component technologies can meet specific goals. Challenges for superconducting computing are reviewed. One challenge is that saving significant energy requires large-scale systems due to the requirement for cryogenic refrigeration, but superconducting computing is a technology in its infancy. For example, in 2010 there were a very few foundries for superconducting electronic circuits and only one with very limited access was processing 200 mm wafers. There were also few superconducting digital circuit designers and the electronic design automation (EDA) toolchain could only support development of circuits with a few thousand elements. These challenges are shared with several other alternative technologies being developed for energy-efficient computing. The IARPA Cryogenic Computing Complexity (C3) program is in the process of developing superconducting computing technologies for large-scale applications.
Special Section on Progress & Trend of Superconductor-based ComputersIt is our great pleasure to have a Special Section entitled "Progress & Trend of Superconductor-based Computers" in IEICE Transactions on Electronics.Superconducting devices based on Josephson junctions exhibit low power consumption and fast switching.Superconducting transmission lines are free of dispersion up to several hundred GHz.Unique logic circuits such as fast single flux quantum (SFQ) circuits and ultralow power adiabatic quantum-flux parametron (AQFP) logic circuits were proposed and have been improved.These circuits have been well developed based on not only their theoretical advantage but also reliable design and fabrication technology of superconducting integrated circuits.Recently, strong interest on quantum computing motivates the development of superconducting quantum-bits, control and readout circuits, and their integration.In this Special Section, seven invited papers and a contributed one describe these topics: two papers for SFQ, four for AQFP, and two for key devices for quantum circuits.Finally, I would like to appreciate activities of all authors, reviewers, and members in editorial committee for completing this special section.
Large-scale computing system characteristics vary by application class, but power and energy use has become a major problem for all classes. Superconducting computing may be able to serve the needs of these systems significantly better than conventional technology. Recent developments in single flux quantum circuit technology for digital logic include variants with greatly improved energy efficiency. Concepts were investigated for computing systems capable of performance in the range from 1 to 1000 PFLOP/s. The concept systems were constrained to use existing commercial cryogenic refrigerators and Nb superconducting technology. In order to meet the performance goals, cache and main memory capable of operating at cryogenic temperatures will be required. Superconducting computing is shown to be potentially competitive on the basis of power and energy efficiency if key component technologies can meet specific goals. Potential advantages of superconducting computing are identified as well as areas requiring further development.