A novel on-chip delay measurement hardware for efficient speed-binning

As technology scales to 45 nm and below, process variations will present significant impact on path delay. This trend makes the deviation between simulated path delay and actual path delay in a manufactured chip more significant. In this paper, we propose a new on-chip path delay measurement structure called path-based ring oscillator (Path-RO). The proposed method creates an oscillator from a targeted path for which it is used to measure path delay on-chip under the impact of process variations. To alleviate accuracy degradation caused by the architecture itself, a high-accuracy calibration process is presented. Through experimental results on Path-ROs inserted in ITCpsila99 b19 benchmark, we obtain path delay distribution under different process variations. The accuracy and efficiency of path delay measurement using Path-RO are also verified by comparing the results obtained from post-layout Hspice simulations.

Dynamic voltage scaling (DVS) provides power savings for systems with varying performance requirements. One low overhead implementation of DVS uses PMOS power switches to connect DVS blocks to one of the available VDD supplies. While power switches have been analyzed extensively for leakage power gating, proper design of power switches for DVS is less well understood. This paper characterizes power switches for DVS in terms of VDD-switching delay and VDD-switching energy. We show the impact of these switching overheads on a novel fine-grained DVS architecture and present an RC model that allows fast estimation of the overhead. Measurements of a DVS multiplier and adder on a 90 nm CMOS test chip validate the model. Our model and measurements confirm that power switched DVS can provide sufficiently low overhead to give energy savings with only one clock cycle spent at a lower voltage, making this approach a flexible and enticing option for embedded portable systems.

This paper discusses a statistical effect of performance optimization to uncertainty in circuit delay. Performance optimization has an effect of balancing the delay of each path in a circuit, i.e., the delay of long paths are shortened and the delay of short paths are lengthened. In these path-balanced circuits, the uncertainty in circuit delay, which are caused by delay calculation error, manufacturing variability, fluctuation of operating condition, etc., becomes worse by a statistical characteristic of delay. Thus, a highly-optimized circuit may not satisfy delay constraints. In this paper, we demonstrate some examples that uncertainty in circuit delay is increased by path-balancing, and we then raise a problem that performance optimization increases statistically-distributed circuit delay.

Conventional voltage scaling systems require a delay margin to maintain a certain level of robustness across all possible device and wire process variations and temperature fluctuations. This margin is required to cover for a possible change in the critical path due to such variations. Moreover, a slower interconnect delay scaling with voltage compared to logic delay can cause the critical path to change from one operating voltage to another. With technology scaling, both process variation and interconnect delay are growing and demanding more margin to guarantee an error-free operation. Such margin is translated into a voltage overhead and a corresponding energy inefficiency. In this paper, a critical path emulator architecture is shown to track the changing critical path at different process splits by probing the actual transistor and wire conditions. Furthermore, voltage scaling characteristics of the actual critical path is closely tracked by programming logic and interconnect delay lines to achieve the same delay combination as the actual critical path. Compared to conventional open-loop and closed-loop systems, the proposed system is up to 39% and 24% more energy efficient, respectively. A 0.18-mum technology test chip is designed to verify the functionality of the proposed system showing critical path tracking of a 16times16 bit multiplier

With the scaling of CMOS technology, critical paths in digital circuits have become largely sensitive to process, voltage and temperature variations as well as to aging effects, generally resulting into a mismatch between the simulated path delay of the circuit obtained with CAD tools and the actual path delay on the manufactured chip. In order to solve this issue and to also avoid conservative strategies based on increasing time margins, adaptive techniques are the most desirable solution because they should automatically sense and correct timing variations online. Implementing such adaptive strategies requires accurate, high resolution and compact delay measurement devices. In this work we propose an effective, fully-digital, online delay measurement circuit that can be entirely implemented in a standard cell technology without the need of custom elements. Our design provides low-cost multi-paths delay monitoring while achieving high accuracy of the measurements (in the order of 30ps).

This paper describes a combination transistor sizing/layout compaction tool used to synthesize high performance CMOS circuits. This optimization tool is part of a large integrated layout system, called Tailor. Given any CMOS circuit layout, Tailor's transistor size optimizer will simultaneously adjust transistor sizes and compact the layout so that the minimum required area (cell pitch) for a specified upper bound on circuit delay is achieved. All delay paths are considered by modeling circuit delay with a logic independent delay graph. Tailor's optimizer globally optimizes circuit area (in one dimension) and delay by use of compaction and nonlinear programming algorithms. The optimizer does not yet optimize in two dimensions simultaneously or optimize hierarchical circuits. Results for a few optimized CMOS circuits are presented.

In this paper, we present a novel on-chip path delay measurement architecture for efficiently detecting and debugging of delay faults in the fabricated integrated circuits. Several delay stages are employed in the proposed on-chip path delay measurement (OCDM) circuit, whose delay ranges are increased by a factor of two gradually from the last to the first delay stage. Thus, the proposed OCDM circuit can achieve a large delay measurement range with a small quantity of delay stages. A calibration circuit is incorporated into the proposed on-chip path delay measurement technique to calibrate the delay range of the delay stage under process variations. In addition, delay calibration for import lines is conducted to improve the precision of path delay measurement. Experimental results are presented to validate the proposed path delay measurement architecture.

The problem of minimizing dynamic power consumption by scaling down the supply voltage of computational elements off critical paths is widely addressed in the literature for the case of combinational designs. The problem is NP-hard in general. To address this problem in the case of synchronous sequential digital designs, one needs to move some registers while applying voltage scaling. Moving these registers shifts some computational elements from critical paths, and can be done by basic retiming. Integrating basic retiming and voltage scaling to address this NP-hard problem cannot in general be done in polynomial run-time. We propose to first apply a guided retiming and then to apply supply voltage scaling on the retimed design. We devise new polynomial time algorithms to realize this guided retiming, and the supply voltage scaling on the retimed design. Experimental results on known benchmarks have shown that the proposed approach can reduce dynamic power consumption by factors as high as 61% for single-phase designs with minimal clock period. Also, they have shown that it can solve optimally the problem, and produce converter-free designs with reduced dynamic power consumption. For large size circuits from ISCAS'89 benchmark suite, the proposed algorithms run in 15s-1h.

Although users of high-performance computing are most interested in raw performance, both energy and power consumption have become critical concerns. Some microprocessors allow frequency and voltage scaling, which enables a system to reduce CPU performance and power when the CPU is not on the critical path. When properly directed, such dynamic frequency and voltage scaling can produce significant energy savings with little performance penalty. This paper presents an MPI runtime system that dynamically reduces CPU performance during communication phases in MPI programs. It dynamically identifies such phases and, without profiling or training, selects the CPU frequency in order to minimize energy-delay product. All analysis and subsequent frequency and voltage scaling is within MPI and so is entirely transparent to the application. This means that the large number of existing MPI programs, as well as new ones being developed, can use our system without modification. Results show that the average reduction in energy-delay product over the NAS benchmark suite is 10% - the average energy reduction is 12% while the average execution time increase is only 2.1%

Dynamic Voltage and Frequency Scaling (DVFS) technique requires accurate observation of critical path delay for robust operation under aggressive supply voltage scaling. Logic paths contain several types of logic gates and path delay have voltage dependences because different logic gates have different voltage dependences. However, it is not well investigated that how the voltage dependence of the path delay changes induced by process variation. This paper describes the effect of the process variation on the voltage dependence of path delay. Ring Oscillator circuits fabricated in 65-nm CMOS process are used for the evaluation and analysis of the process variation dependence of the voltage delay curves.

Power/energy reduction is of uttermost importance for applications with stringent power/energy budget such as ultra-low power and energy-harvested systems. Aggressive voltage scaling and in particular Near-Threshold Computing (NTC) is a promising approach to reduce the power and energy consumption. However, reducing the supply voltage leads to drastic performance variation induced by process and runtime variation. Temperature variation is one of the major sources of performance variation. In this paper, we study the impact of temperature variation on the circuit behavior at near threshold region and show that the ambient temperature has a huge impact on the metrics such as circuit delay, power and energy consumption. We also propose a low-cost, ambient temperature-aware voltage scaling technique to reduce the unnecessary energy overhead caused by temperature variation. Simulation results show that our proposed approach reduces the energy consumption by more than 3X.

As technology shrinks, gate sensitivity to noise increases due to supply voltage scaling and limited scaling of the voltage threshold. As a result, power supply noise (PSN) plays a greater role in sub-100 nm technologies and creates signal integrity issues. It is vital to consider supply voltage noise effects: 1) during design validation to apply sufficient guardbands to critical paths, and 2) during path delay test to ensure the performance and reliability of the chip. In this paper, a novel layout-aware pattern generation procedure is proposed to maximize PSN effects on critical paths considering the impact of local voltage drop. The proposed pattern generation and validation flow is implemented on the ITC'99 b19 benchmark. Experimental results for both wire-bond and flip-chip packaging styles are presented. Results demonstrate that our proposed method is fast, significantly increases switching around the functionally testable critical paths, and induces large voltage drop on cells placed on the critical paths which results in increased path delay. The proposed method eliminates the very time consuming pattern validation phase that is practised in industry.

A new field programmable gate array (FPGA) design algorithm, Maple-opt, is proposed for technology mapping, placement, and global routing subject to a given upper bound of critical signal path delay. The basic procedure of Maple-opt is viewed as top-down hierarchical bipartition of a layout region. In each bipartitioning step, technology mapping onto logic blocks of FPGAs, their placement, and global routing are determined simultaneously, which leads to a more congestion-balanced layout for routing. In addition, Maple-opt is capable of estimating a lower bound of the delay for a constrained path and of extracting critical paths based on the difference between the lower bounds and given constraint values in each bipartitioning step. Two delay-reduction procedures for the critical paths are applied; routing delay reduction and logic-block delay reduction. The routing delay reduction is done by assigning each constrained path to a single subregion when bipartitioning a region. The logic-block delay reduction is done by mapping each constrained path onto a smaller number of logic blocks. Experimental results for benchmark circuits demonstrate that Maple-opt reduces the maximum number of tracks per channel by a maximum of 38% compared with existing algorithms while satisfying almost all the path delay constraints.

Today's microprocessors and Systems-on-Chip are thermally limited. Many, therefore, employ dynamic thermal management (DTM) to maximize performance under a reliability constraint. Accurate thermal monitoring is critical as temperature underestimation can hurt reliability by excessively aging devices and overestimation can hurt performance by unnecessarily throttling computing components. Placing temperature sensors close to potential hotspots can help accuracy, but it is non-trivial as hotspots often form inside digital blocks consisting of densely placed digital cells. Large sensors can disrupt cell placement thereby increasing wire lengths and circuit delays. Furthermore, the sensor needs to operate from the same digital power grid as the circuit, one that can be scaled down to near-threshold regime via dynamic voltage scaling. Absent this ability, a separate power grid and dedicated supply voltage for the sensors further increases area overhead. In this paper, we present a sensor circuit that is compact and deeply voltage-scalable and can be embedded among digital cells with little disruption. Simulation results show that it achieves a comparable accuracy to other compact sensor circuits for DTM.

This paper presents GRACE-OS, an energy-efficient soft real-time CPU scheduler for mobile devices that primarily run multimedia applications. The major goal of GRACE-OS is to support application quality of service and save energy. To achieve this goal, GRACE-OS integrates dynamic voltage scaling into soft real-time scheduling and decides how fast to execute applications in addition to when and how long to execute them. GRACE-OS makes such scheduling decisions based on the probability distribution of application cycle demands, and obtains the demand distribution via online profiling and estimation. We have implemented GRACE-OS in the Linux kernel and evaluated it on an HP laptop with a variable-speed CPU and multimedia codecs. Our experimental results show that (1) the demand distribution of the studied codecs is stable or changes smoothly. This stability implies that it is feasible to perform stochastic scheduling and voltage scaling with low overhead; (2) GRACE-OS delivers soft performance guarantees by bounding the deadline miss ratio under application-specific requirements; and (3) GRACE-OS reduces CPU idle time and spends more busy time in lower-power speeds. Our measurement indicates that compared to deterministic scheduling and voltage scaling, GRACE-OS saves energy by 7% to 72% while delivering statistical performance guarantees.