CMOS microprocessors are limited in their capacity for clock speed improvement because of increasing computing power, i.e., they face a power-wall problem. Single-flux-quantum (SFQ) circuits offer a solution with their ultra-fast-speed and ultra-low-power natures. This paper introduces our contributions towards ultra-high-speed cryogenic SFQ computing. The first step is to design SFQ microprocessors. From qualitatively and quantitatively evaluating past-designed SFQ microprocessors, we have found that revisiting the architecture of SFQ microprocessors and on-chip caches is the first critical challenge. On the basis of cross-layer discussions and analysis, we came to the conclusion that a bit-parallel gate-level pipeline architecture is the best solution for SFQ designs. This paper summarizes our current research results targeting SFQ microprocessors and on-chip cache architectures.

We describe a large-scale integrated circuit (LSI) design of rapid single-flux-quantum (RSFQ) circuits and demonstrate several reconfigurable data-path (RDP) processor prototypes based on the ISTEC Advanced Process (ADP2). The ADP2 LSIs are made up of nine Nb layers and Nb/AlOx/Nb Josephson junctions with a critical current density of 10kA/cm2, allowing higher operating frequencies and integration. To realize truly large-scale RSFQ circuits, careful design is necessary, with several compromises in the device structure, logic gates, and interconnects, balancing the competing demands of integration density, design flexibility, and fabrication yield. We summarize numerical and experimental results related to the development of a cell-based design in the ADP2, which features a unit cell size reduced to 30-µm square and up to four strip line tracks in the unit cell underneath the logic gates. The ADP LSIs can achieve ∼10 times the device density and double the operating frequency with the same power consumption per junction as conventional LSIs fabricated using the Nb four-layer process. We report the design and test results of RDP processor prototypes using the ADP2 cell library. The RDP processors are composed of many arrays of floating-point units (FPUs) and switch networks, and serve as accelerators in a high-performance computing system. The prototypes are composed of two-dimensional arrays of several arithmetic logic units instead of FPUs. The experimental results include a successful demonstration of full operation and reconfiguration in a 2×2 RDP prototype made up of 11.5k junctions at 45GHz after precise timing design. Partial operation of a 4×4 RDP prototype made up of 28.5k-junctions is also demonstrated, indicating the scalability of our timing design.

In this paper, we propose a method for layout-driven skewed clock tree synthesis for SFQ logic circuits. For a given logic circuit without a clock tree, our algorithm outputs a circuit with a synthesized clock tree and timing adjustments achieving the given clock period and a rough placement of the clocked gates. In the proposed algorithm, clocked gates are grouped into levels and the clock tree is synthesized for each level. For each level, we estimate the clock timing for all possible placements of each gate, and then we search a placement of all gates that minimizes the total number of delay elements for timing adjustment. Once the placement is obtained, we synthesize a clock tree without wire intersections. We applied the proposed method to a moderate size circuit and confirmed that clock trees satisfying given timing requirements can be synthesized automatically.

We present circuit implementations for computing exponentials and logarithms suitable for rapid single-flux-quantum (RSFQ) logic. We propose hardware algorithms based on the sequential table-lookup (STL) method using the radix-2 signed-digit representation that achieve high-throughput, digit-serial calculations. The circuits are implemented by processing elements formed in systolic-array-like, regularly-aligned pipeline structures. The processing elements are composed of adders, shifters, and readouts of precomputed constants. The iterative calculations are fully overlapped, and throughputs approach the maximum throughput of serial processing. The circuit size for calculating significand parts is estimated to be approximately 5-10 times larger than that of a bit-serial floating-point adder or multiplier.

We demonstrated an automated passive-transmission-line routing tool for single-flux-quantum (SFQ) circuits. The tool is based on the A* algorithm, which is widely used in CMOS LSI design, and tuned for microstrip/strip lines formed in the SRL 4-Nb layer structure. In large-scale SFQ circuits with 10000-20000 Josephson junctions, such as microprocessors, 80-90% of the wires can be automatically routed in about ten minutes. We verified correct operation above 40 GHz for an automatically routed 44 switch circuit from on-chip high-speed tests. The resulting circuit size and operating frequency were comparable to those of a manually designed result. We believe that the tool is useful for large-scale SFQ circuit design using conventional fabrication processes.

We present a concept of an advanced rapid single-flux-quantum (RSFQ) logic circuit family using the combination of 0-shifted and π-shifted Josephson junctions. A π-shift in the current-phase relationship can be obtained in several types of Josephson junctions, such as Josephson junctions containing a ferromagnet barrier layer, depending on its thickness and temperature. We use a superconducting quantum interference devices composed of a pair of 0- and π-shifted Josephson junctions (0-π SQUIDs) as a basic circuit element. Unlike the conventional RSFQ logic, bistability is obtained by spontaneous circular currents without using a large superconductor loop, and the state can be flipped by smaller driving currents. These features lead to energy- and/or space-efficient logic gates. In this paper, we show several example circuits where we represent signals by flips of the states of a 0-π SQUID. We obtained successful operation of the circuits from numerical simulation.

A single flux quantum (SFQ) logic cell library has been developed for the 10 kA/cm2 Nb multi-layer fabrication process to efficiently design large-scale SFQ digital circuits. In the new cell library, the critical current density of Josephson junctions is increased from 2.5 kA/cm2 to 10 kA/cm2 compared to our conventional cell library, and the McCumber-Stwart parameter of each Josephson junction is increased to 2 in order to increase the circuit operation speed. More than 300 cells have been designed, including fundamental logic cells and wiring cells for passive interconnects. We have measured all cells and confirmed they stably operate with wide operating margins. On-chip high-speed test of the toggle flip-flop (TFF) cell has been performed by measuring the input and output voltages. The TFF cell at the input frequency of up to 400 GHz was confirmed to operate correctly. Also, several fundamental digital circuits, a 4-bit concurrent-flow shift register and a bit-serial adder have been designed using the new cell library, and the correct operations of the circuits have been demonstrated at high clock frequencies of more than 100 GHz.

We report the successful operation of a low-power arithmetic logic unit (ALU) based on a low-voltage rapid single-flux-quantum (LV-RSFQ) logic circuit, whereby a dc bias current is fed to circuits from lowered constant-voltage sources through small resistors. Both the static and dynamic energy consumptions are reduced because of the reduction in the amplitudes of voltage pulses across the Josephson junctions, with a trade-off of slightly slower switching speeds. The designed bias voltage was set to 0.25mV, which is one-tenth that of our standard RSFQ circuit design. We investigated several issues related to such low-voltage operation, including margins and timing design. To achieve successful operation, we tuned the circuit parameters in the logic gate design and carefully controlled the timing by considering the interference of pulse signals. We show test results for the low-voltage ALU in on-chip high-speed testing. The circuit was fabricated using the AIST Nb/AlOx/Nb Advanced Process with a critical current density of 10kA/cm2. We verified that arithmetic and logical operations were correctly implemented and obtained dc bias margins of 18% at a target clock frequency of 20GHz and achieved a maximum clock frequency of 28GHz with a power consumption of 28µW. These experimental results indicate energy efficiency of 3.6 times that of the standard RSFQ circuit design.

Recent advances of superconducting single-flux-quantum (SFQ) circuit technology make it attractive to investigate computing systems using SFQ circuits, where arithmetic circuits play important roles. In order to develop excellent SFQ arithmetic circuits, we have to design or select their underlying algorithms, called hardware algorithms, from different point of view than CMOS circuits, because SFQ circuits work by pulse logic while CMOS circuits work by level logic. In this paper, we compare implementations of hardware algorithms for addition by synchronous-clocking SFQ circuits. We show that a set of individual bit-serial adders and Kogge-Stone adder are superior to others.

We have studied on thermally assisted nano-structured transistors made of superconductor ultra-thin films. These transistors potentially work as interface devices for Josephson-CMOS (complementary metal oxide semiconductor) hybrid memory systems, because they can generate a high output voltage of sub-V enough to drive a CMOS transistor. In addition, our superconductor transistors are formed with very fine lines down to several tens of nm in widths, leading to very small foot print enabling us to make large capacity hybrid memories. Our superconductor transistors are made with niobium titanium nitride (NbTiN) thin films deposited on thermally-oxidized silicon substrates, on which other superconductor circuits or semiconductor circuits can be formed. The NbTiN thickness dependence of the critical temperature and of resistivity suggest thermally activated vortex or anti-vortex behavior in pseudo-two-dimensional superconducting films plays an important role for the operating principle of the transistors. To show the potential that the transistors can drive MOS transistors, we analyzed the driving ability of the superconductor transistors with HSPICE simulation. We also showed the turn-on behavior of a MOS transistor used for readout of a CMOS memory cell experimentally. These results showed the high potential of superconductor transistors for Josephson-CMOS hybrid memories.

We describe the development of single-flux-quantum (SFQ) microprocessors and the related technologies such as designing, circuit architecture, microarchitecture, etc. Since the microprocessors studied here aim for a general-purpose computing system, we employ the complexity-reduced (CORE) architecture in which the high-speed nature of the SFQ circuits is used not for increasing processor performance but for reducing the circuit complexity. The bit-serial processing is the most suitable way to realize the CORE architecture. We assembled all the best technologies concerning SFQ integrated circuits and designed the SFQ microprocessors, CORE1α, CORE1β, and CORE1γ. The CORE1β was made up of about 11000 Josephson junctions and successfully demonstrated. The peak performance reached 1400 million operations per second with a power consumption of 3.4 mW. We showed that the SFQ microprocessors had an advantage in a performance density to semiconductor's ones, which lead to the potential for constructing a high performance SFQ-circuit-based computing system.

We developed a Nb 4-layer process for fabricating superconducting integrated circuits that involves using caldera planarization to increase the flexibility and reliability of the fabrication process. We call this process the planarized high-speed standard process (PHSTP). Planarization enables us to flexibly adjust most of the Nb and SiO2 film thicknesses; we can select reduced film thicknesses to obtain larger mutual coupling depending on the application. It also reduces the risk of intra-layer shorts due to etching residues at the step-edge regions. We describe the detailed process flows of the planarization for the Josephson junction layer and the evaluation of devices fabricated with PHSTP. The results indicated no short defects or degradation in junction characteristics and good agreement between designed and measured inductances and resistances. We also developed single-flux-quantum (SFQ) and adiabatic quantum-flux-parametron (AQFP) logic cell libraries and tested circuits fabricated with PHSTP. We found that the designed circuits operated correctly. The SFQ shift-registers fabricated using PHSTP showed a high yield. Numerical simulation results indicate that the AQFP gates with increased mutual coupling by the planarized layer structure increase the maximum interconnect length between gates.

One of the fundamental problems in many-pixel detectors implemented in cryogenics environments is the number of bias and read-out wires. If one targets a megapixel range detector, number of wires should be significantly reduced. One possibility is that the detectors are serially connected and biased by using only one line and read-out is accomplished by on-chip circuitry. In addition to the number of pixels, the detectors should have fast response times, low dead times, high sensitivities, low inter-pixel crosstalk and ability to respond to simultaneous irradiations to individual pixels for practical purposes. We have developed an equivalent circuit model for a serially connected superconducting strip line detector (SSLD) array together with the read-out electronics. In the model we take into account the capacitive effects due to the ground plane under the detector, effects of the shunt resistors fabricated under the SSLD layer, low pass filters placed between the individual pixels that enable individual operation of each pixel and series resistors that prevents the DC bias current flowing to the read-out electronics as well as adjust the time constants of the inductive SSLD loop. We explain the results of investigation of the following parameters: Crosstalk between the neighbor pixels, response to simultaneous irradiation, dead times, L/R time constants, low pass filters, and integration with the SFQ front-end circuit. Based on the simulation results, we show that SSLDs are promising devices for detecting a wide range of incident radiation such as neurons, X-rays and THz waves in many-pixel configurations.