Pipelining of high performance synchronous digital systems

A new approach is described to the design of those synchronous digital systems in which the performance parameters latency and clock frequency are of primary importance. Specifically, the trade-off between clock frequency and latency is analysed in terms of the circuit characteristics of a pipelined data path. A design paradigm relating latency and clock frequency as a function of the level of pipelining is described for studying the performance of a synchronous system. This perspective permits the development of design equations for constrained and unconstrained design problems from which the optimal level of pipelining can be determined in terms of the delays of logic, interconnect, and registers, and the clock skew and number of logic stages.     

Globally updated mesochronous design style

In large-scale and high-speed digital systems, global synchronization has frequently been used to protect clocked I/O from data failure due to metastability. Synchronous design styles are widely used, easy to grasp and to implement, and also well supported by logic synthesis tools. There are many drawbacks with global synchronization. Most important is the relationship between physical size and maximum clock frequency, which will approach its limit as clock frequency and system size increase simultaneously. The purpose of this proposed Globally Updated Mesochronous (GUM) design style is to overcome those drawbacks by identifying all global signal links in the system and adding synchronization circuits to these. System level simplicity, inherited from synchronous design and its tool support, is retained. In this paper, the GUM design style is described, analyzed, and demonstrated. Experimental results from a large-scale high-speed system using three 0.8-/spl mu/m BiCMOS chips are given. The GUM design style is scaleable and suitable for future system-on-chip applications both on and among chips.     

Superconducting digital electronics

Single-flux quantum logic (SFQ) circuits, in which a flux quantum is used as an information carrier, have the possibility for opening the door to a new digital system operated at over 100-GHz clock frequency at extremely low power dissipation. The SFQ logic system is a so-called pulse logic, which is completely different from the level logic for semiconductors like CMOS, so circuit design technologies for SFQ logic circuits have to be newly developed. Recently, much progress in basic technologies for designing SFQ circuits and operating circuits at high speeds has been made. With advances in these design tools, large-scale circuits including more than several thousand junctions can be easily operated with the clock frequency of more than several tens of gigahertz. High-end routers and high-end computers are possible applications of SFQ logic circuits because of their high throughput nature and the low power dissipation of SFQ logic. In this paper, recent advances of SFQ circuit design technologies and recent developments of switches for high-end routers and microprocessors for high-end computers that are considered possible applications for SFQ logic will be described.     

A gallium-arsenide digital phase shifter for clock and controlsignal distribution in high-speed digital systems

The design, implementation, testing, and applications of a gallium-arsenide digital phase shifter and fan-out buffer are described. The integrated circuit provides a method for adjusting the phase of high-speed clock and control signals in digital systems, without the need for pruning cables, multiplexing between cables of different lengths, delay lines, or similar techniques. The phase of signals distributed with the described chip can be dynamically adjusted in eight different steps of approximately 60 ps per step. The IC also serves as a fan-out buffer and provides 12 in-phase outputs. The chip is useful for distributing high-speed clock and control signals in synchronous digital systems, especially if components are distributed over a large physical area or if there is a large number of components     

Performance analysis of a complete adiabatic logic system driven by the proposed power clock generator

We analyze the energy performance of a complete adiabatic circuit/system including the Power Clock Generator （PCG） at the 90 nm CMOS technology node. The energy performance in terms of the conversion efficiency of the PCG is extensively carried out under the variations of supply voltage, process comer and the driver transistor＇s width. We propose an energy-efficient singe cycle control circuit based on the two-stage comparator for the synchronous charge recovery sinusoidal power clock generator （PCG）. The proposed PCG is used to drive the 4-bit adiabatic Ripple Carry Adder （RCA） and their simulation results are compared with the adiabatic RCA driven by the reported PCG. We have also simulated the logically equivalent static CMOS RCA circuit to compare the energy saving of adiabatic and non-adiabatic logic circuits. In the clock frequency range from 25 MHz to 1GHz, the proposed PCG gives a maximum conversion efficiency of 56.48%. This research work shows how the design of an efficient PCG increases the energy saving of adiabatic logic.     

Clock jitter generator with picoseconds resolution

The clock is one of the most critical signals in any synchronous system. As CMOS technology has scaled, supply voltages have dropped chip power consumption has increased and the effects of jitter due to clock frequency increase have become critical and jitter budget has become tighter. This article describes design and development of low-cost mixed-signal programmable jitter generator with high resolution. The digital technique is used for coarse-grain and an analogue technique for fine-grain clock phase shifting. Its structure allows injection of various random and deterministic jitter components in a controllable and programmable fashion. Each jitter component can be switched on or off. The jitter generator can be used in jitter tolerance test and jitter transfer function measurement of high-speed synchronous digital circuits. At operating system clock frequency of 220?MHz, a jitter with 4?ps resolution can be injected.     

High Performance, Energy Efficiency, and Scalability With GALS Chip Multiprocessors

Chip multiprocessors with globally asynchronous locally synchronous (GALS) clocking styles are promising candidates for processing computationally-intensive and energy-constrained workloads. The GALS methodology simplifies clock tree design, provides opportunities to use clock and voltage scaling jointly in system submodules to achieve high energy efficiencies, and can also result in easily scalable clocking systems. However, its use typically also introduces performance penalties due to additional communication latency between clock domains. We show that GALS chip multiprocessors (CMPs) with large inter-processor first-inputs–first-outputs (FIFOs) buffers can inherently hide much of the GALS performance penalty while executing applications that have been mapped with few communication loops. In fact, the penalty can be driven to zero with sufficiently large FIFOs and the removal of multiple-loop communication links. We present an example mesh-connected GALS chip multiprocessor and show it has a less than 1% performance (throughput) reduction on average compared to the corresponding synchronous system for many DSP workloads. Furthermore, adaptive clock and voltage scaling for each processor provides an approximately 40% power savings without any performance reduction. These results compare favorably with the GALS uniprocessor, which compared to the corresponding synchronous uniprocessor, has a reported greater than 10% performance (throughput) reduction and an energy savings of approximately 25% using dynamic clock and voltage scaling for many general purpose applications.      

Power Management of Voltage/Frequency Island-Based Systems Using Hardware-Based Methods

Shrinking technology nodes combined with the need for higher clock speeds have made it increasingly difficult to distribute a single global clock across a chip while meeting the power requirements of the design. Globally asynchronous locally synchronous (GALS) design style can help achieve low power consumption and modularity of a design while greatly reducing the number of global interconnects. Such multiple clock domain architectures can benefit from having frequency/voltage values assigned to each domain based on workload requirements. The work presented in this paper proposes a new hardware-based approach to dynamically change the frequencies and potentially voltages of a voltage-frequency island (VFI) system driven by a dynamic workload. This technique tries to change the frequency of a synchronous island such that it will have efficient power utilization while satisfying performance constraints. In recent years, there have been major developments, both in industry and academia, in the field of multiprocessor systems. Such multiprocessor systems are very good candidates for VFI design style implementation, where one or more processors can be part of a single VFI. To demonstrate the feasibility of our proposed method, we have implemented a multiprocessor system for a field-programmable gate array (FPGA) platform that uses independently generated clocks for each processor. The results from the FPGA platform confirm the claim that the power consumption of a system can potentially be reduced while maintaining the performance of many applications. Our work concentrates primarily on embedded systems, but the idea can be explored for general-purpose computing as well.      

基于延迟锁相环和锁频环结构的全数字同步倍频器

针对现有基于PLLs/DLLs的全数字化同步倍频器结构存在的不足,本文提出了基于一种双环结构的全数字同步倍频器。它由延迟锁相环和锁频环共享一个共同的参考时钟信号(FREF)构成,不需要任何模拟组件。它可以采用Verilog-HDL语言设计,可在Altera DE2-70开发板上实现合成,而且可以很容易地适应于不同的FPGA系列以及作为一个集成电路实现,同时也可用于为分布式数字处理系统以及片上系统的片内/片间通信提供时钟参考;实验结果表明,本文所提出的结构相比于现有的结构,能够获得更高频率的输出时钟信号,提供更好的频率分辨率、更好的抖动性能和高倍乘因子。     

Design and performance of multistage GaAs dynamic logic

GaAs Two-Phase Dynamic FET Logic (TDFL) circuits are capable of extremely low power dissipation (20 nW/MHz/gate), high speed (1 GHz), and are compatible with static GaAs logic families. This paper demonstrates that TDFL can be modified to execute two or three stages of logic in one clock phase. This extension provides extremely high functional complexity per gate that can be used to reduce power dissipation, reduce latency, and increase circuit density in both sequential and computationally-oriented applications. The performance of these gates was demonstrated by E/D MESFET IC test circuits fabricated by a digital IC foundry. A one clock cycle, 8-b carry-lookahead adder operated at 350 MHz with only 1.1 mW of power dissipation     

Short-term frequency stability of clock generators for multigigahertz rapid-single-flux quantum digital circuits

Superconducting digital systems based on Josephson junctions have generally used a synchronous timing strategy. A master clock signal is used to delimit a data window during which the system changes state and data is transferred from one block to the next. The temporal stability of the clock signal has a profound effect on the performance of rapid single flux quantum (RSFQ) digital systems. In particular, short-term clock fluctuations, or clock jitter, can degrade system performance due to the hazard of timing constraint violations. The successful development of large-scale RSFQ digital systems will require highly stable multigigahertz on-chip clock sources. To meet this need, methods for characterizing and measuring the short-term stability of such sources are required. We identify the relevant figure of merit to characterize and compare various clocks: the cycle-to-cycle standard deviation of the clock periods. We present experimental techniques for the measurement of this figure of merit and apply them to the measurement of jitter in a clock generator used often in RSFQ systems, the ring oscillator. High-frequency phase noise measurements found the jitter of a 9.6-GHz clock to be in the range from 0.6% to 0.36% of the clock period. The measured values of clock jitter fell within the 95% confidence interval of our stochastic circuit simulations. This was sufficient evidence to conclude that thermal noise from the resistors in the circuit may be the dominant source of jitter in the ring oscillator.     

P2R2: Parallel Pseudo-Round-Robin arbiter for high performance NoCs

Networks-on-Chip (NoCs) play an important role in the performance of Chip Multi-Processors (CMPs). Providing the desired performance under heavy traffics imposed by some applications necessitates NoC routers to have a large number of Virtual Channels (VCs). Increasing the number of VCs, however, will add to the delay of the critical path of the arbitration logic, and hence restricts the clock frequency of the router. In order to make it possible to enjoy the benefits of having many VCs and keep the clock frequency as high as that of a low-VC router, we propose Parallel Pseudo-Round-Robin (P2R2) arbiter. Our proposal is based on processing multiple groups of requests in parallel. Our experimental results show that the proposed scheme can beat the state-of-the-art arbiter design by up to 12.5% and 6.8% in terms of saturation rate and zero-load latency, respectively, under synthetic traffic patterns. These results also demonstrate a 29.5% improvement in average packet latency in Splash-2 applications in favor of P2R2 with respect to the state-of-the-art arbiter.     

Timing of Multi-Gigahertz Rapid Single Flux Quantum Digital Circuits

Rapid Single Flux Quantum (RSFQ) logic is a digital circuit technology based on superconductors that has emerged as a possible alternative to advanced semiconductor technologies for large scale ultra-high speed, very low power digital applications. Timing of RSFQ circuits at frequencies of tens to hundreds of gigahertz is a challenging and still unresolved problem. Despite the many fundamental differences between RSFQ and semi- conductor logic at the device and at the circuit level, timing of large scale digital circuits in both technologies is principally governed by the same rules and constraints. Therefore, RSFQ offers a new perspective on the timing of ultra-high speed digital circuits.This paper is intended as a comprehensive review of RSFQ timing, from the viewpoint of the principles, concepts, and language developed for semiconductor VLSI. It includes RSFQ clocking schemes, both synchronous and asynchronous, which have been adapted from semiconductor design methodologies as well as those developed specifically for RSFQ logic. The primary features of these synchronization schemes, including timing equations, are presented and compared.In many circuit topologies of current medium to large scale RSFQ circuits, single-phase synchronous clocking outperforms asynchronous schemes in speed, device/area overhead, and simplicity of the design procedure. Synchronous clocking of RSFQ circuits at multigigahertz frequencies requires the application of non-standard design techniques such as pipelined clocking and intentional non-zero clock skew. Even with these techniques, there exist difficulties which arise from the deleterious effects of process variations on circuit yield and performance. As a result, alternative synchronization techniques, including but not limited to asynchronous timing, should be considered for certain circuit topologies. A synchronous two-phase clocking scheme for RSFQ circuits of arbitrary complexity is introduced, which for critical circuit topologies offers advantages over previous synchronous and asynchronous schemes.     

Electrical and optical clock and data recovery in optical access networks: a comparative study

Clock and data recovery (CDR) is an essential part in high‐speed telecommunication systems. The CDR is used to extract the clock and re‐time the received data, which allows a synchronous operation to recover the transmitted signal. In optical access networks, electrical CDR or optical CDR implementations can be used. However, there are no clear guidelines or recommendations on which CDR implementation should be adopted for better performance. These missing clear recommendations are because the electrical CDR requires electronics design expertise whereas the optical CDR requires optical design expertise. Consequently, in this paper, an all‐digital CDR, designed and implemented on the field‐programmable gate array platform, and an optical CDR, developed by using fiber Bragg grating technology on the OptiSystem platform, are presented. Furthermore, the integration of these 2 CDR implementations with the optical access network is implemented, and their performance is evaluated for various transmission rates and communication distances. Finally, a comparative study in terms of the bit error rate between the all‐digital CDR and the optical CDR is presented.     

A "flying-adder" architecture of frequency and phase synthesis with scalability

Most of today's digital designs, from small-scale digital block designs to system-on-chip (SoC) designs, are based on "synchronous" design principle. Clock is the most important issue in these designs. Frequency and phase synthesis is closely related to the clock generation. A frequency and phase synthesis technique based on phase-locked loop is proposed in that delivers high performance, easy integration, and high stability. However, there are problems associated with this architecture, such as: 1) its highest deliverable frequency is limited by the speed of the accumulator and 2) the phase synthesis circuitry will not work well in certain ranges (dead zone) and in certain conditions (dual stability). This paper presents an improved architecture that addresses these problems. The new frequency synthesis circuitry has scalability for higher output frequency. It also has an internal node whose frequency is twice that of output signal. When duty cycle is not a concern, this signal can be used directly as clock source. The new phase synthesis circuitry is free of "dead zone" and "dual stability." The improved architecture has better performance, is simpler to implement, and is easier to understand.     

多模式数字同步时钟产生技术研究

同步时钟是网络同步的关键，本文对基于多模式数字同步时钟产生的系统构成、关键技术、关键技术对系统性能的影响等进行了研究。其中多模式同步时钟源采用来自GPS／GLONASS，本地时钟、‘北斗’一号卫星定时接收嚣、BPM授时短波接收嚣和本地时钟，经过输入处理和表决判优，选取最优时钟信号作为同步时钟源。     

异步式逻辑保障片上网络传输可靠性的研究

随着Si技术的持续发展,片上系统(SoC)的规模和复杂度的增长给传统的片上互连,如总线结构,带来了前所未有的挑战。片上网络[1-2]是片上系统的一种新设计方法,是目前公认应对这种挑战较为有效的解决方案。半导体工艺进入深亚微米时代后,片上网络的可靠性也越来越成为人们关注的问题。将在研究如何应用异步式逻辑保障片上网络互连数据传输的可靠性和服务质量,提出了一个异步式片上网络的架构。通过实验证明,异步式逻辑将极大提高集成电路在应对电源不稳定性、导线间串扰、电磁干扰(EMI)、时钟偏斜和软错误方面的可靠性。采用全局异步局部同步的时钟机制,该方法带来了一种全新的片上通信方法,显著改善了传统总线式系统的性能。     

NRZ伪随机码序列同步时钟提取

在光纤数字通信系统中,ＮＲＺ非归零码是一种最常见的基带信号。它本身不含有位同步时钟分量。只有对其进行非线性处理,转换成ＲＺ归零码后,方可提取出同步时钟,本文通过对ＮＲＺ,ＲＺ伪随机码序列进行频谱分析,得知当ＮＲＺ码变换成码元占空比为１／２的ＲＺ码时,所提取的同步时钟功率最强。     

Clock distribution networks in synchronous digital integratedcircuits

Clock distribution networks synchronize the flow of data signals among synchronous data paths. The design of these networks can dramatically affect system-wide performance and reliability. A theoretical background of clock skew is provided in order to better understand how clock distribution networks interact with data paths. Minimum and maximum timing constraints are developed from the relative timing between the localized clock skew and the data paths. These constraint relationships are reviewed, and compensating design techniques are discussed. The field of clock distribution network design and analysis can be grouped into a number of subtopics: 1) circuit and layout techniques for structured custom digital integrated circuits; 2) the automated layout and synthesis of clock distribution networks with application to automated placement and routing of gate arrays, standard cells and larger block-oriented circuits; 3) the analysis and modeling of the timing characteristics of clock distribution networks; and 4) the scheduling of the optimal timing characteristics of clock distribution networks based on architectural and functional performance requirements. Each of these areas is described the clock distribution networks of specific industrial circuits are surveyed and future trends are discussed     

一种基于标准逻辑单元的GALS异步封装电路

基于点对点GALS模型,给出了异步封装电路的信号状态转换图(STG),基于Petrify设计了一种基于标准逻辑单元的GALS异步封装电路,包括同步/异步接口电路、具有分频及暂停功能的局部时钟等设计.由于所设计的异步封装电路具有不存在延时器件、没有使用特殊的异步逻辑单元等特点,所以论文基于两个同步计数器实现了GALS点对点模型进行仿真和FPGA验证,结果显示了整个异步封装及其GALS系统性能的正确性.