Search

Deep neural networks (DNNs) are being heavily utilized in modern applications, putting energy-constraint devices to the test. To bypass high energy consumption issues, approximate computing has been employed in DNN accelerators to balance out the accuracy-energy reduction trade-off. However, the approximation-induced accuracy loss can be very high and drastically degrade the performance of the DNN. Therefore, there is a need for a fine-grain mechanism that would assign specific DNN operations to approximation to maintain acceptable DNN accuracy, while achieving low energy consumption. We present an automated framework for weight-to-approximation mapping through formal property exploration for approximate DNN accelerators. At the MAC unit level, our experimental evaluation surpassed already energy-efficient mappings by more than $\times 2$ in terms of energy gains, while supporting a fine-grain control over the introduced approximation. [ABSTRACT FROM AUTHOR]

Recent breakthroughs in Neural Networks (NNs) have made DNN accelerators ubiquitous and led to an ever-increasing quest on adopting them from Cloud to edge computing. However, state-of-the-art DNN accelerators pack immense computational power in a relatively confined area, inducing significant on-chip power densities that lead to intolerable thermal bottlenecks. Existing state of the art focuses on using approximate multipliers only to trade-off efficiency with inference accuracy. In this work, we present a thermal-aware approximate DNN accelerator design in which we additionally trade-off approximation with temperature effects towards designing DNN accelerators that satisfy tight temperature constraints. Using commercial multi-physics tool flows for heat simulations, we demonstrate how our thermal-aware approximate design reduces the temperature from 139 $^{\circ }$ ∘ C, in an accurate circuit, down to 79 $^{\circ }$ ∘ C. This enables DNN accelerators to fulfill tight thermal constraints, while still maximizing the performance and reducing the energy by around 75% with a negligible accuracy loss of merely 0.44% on average for a wide range of NN models. Furthermore, using physics-based transistor aging models, we demonstrate how reductions in voltage and temperature obtained by our approximate design considerably improve the circuit’s reliability. Our approximate design exhibits around 40% less aging-induced degradation compared to the baseline design. [ABSTRACT FROM AUTHOR]

Recent breakthroughs in Neural Networks (NNs) led to significant accuracy improvements of several machine learning applications such as image classification and voice recognition. However, this accuracy improvement comes at the cost of an immense increase in computation demands. NNs became one of the most common and computationally intensive workloads in today's datacenters. To address these computational demands, Google announced in 2016 the Tensor Processing Unit (TPU), an advanced custom ASIC accelerator for NN inference. Two new TPU versions (v2 and v3) followed in 2017 and 2018 that support also training. Google TPUv3 packs an immense processing power ($\mathrm{90TFLOPS}$ 90 TFLOPS per chip) in a tiny and condensed area, leading to very high on-chip power densities and thus excessive temperature. In this article, superlattice thermoelectric cooling, which is one of the emerging on-chip cooling, is considered as an advanced cooling example for Google TPU and we investigate the impact of Negative Capacitance FET (NCFET), which is one of the recent emerging technologies, on the cooling and efficiency of TPU. Through full-chip design, of the computational core of the TPU, based on $14\mathrm{nm}$ 14 nm Intel FinFET technology and multiphysics temperature simulations, we demonstrate that NCFET can significantly minimize the required cooling-cost. More than 4000 NCFET configurations are evaluated in order to traverse the entire design space defined by the thickness of the ferroelectric layer of NCFET, the operating voltage, cooling, and the operating frequency, in addition to all possible FinFET's configurations. Moreover, our experimental evaluation shows that by eliminating the cooling cost, NCFET delivers 2.8x higher efficiency compared to the conventional FinFET baseline. [ABSTRACT FROM AUTHOR]

Neural processing units (NPUs) are becoming an integral part in all modern computing systems due to their substantial role in accelerating neural networks (NNs). The significant improvements in cost-energy-performance stem from the massive array of multiply accumulate (MAC) units that remarkably boosts the throughput of NN inference. In this work, we are the first to investigate the thermal challenges that NPUs bring, revealing how MAC arrays, which form the heart of any NPU, impose serious thermal bottlenecks to on-chip systems due to their excessive power densities. For the first time, we explore: 1) the effectiveness of precision scaling and frequency scaling (FS) in temperature reductions and 2) how advanced on-chip cooling using superlattice thin-film thermoelectric (TE) open doors for new tradeoffs between temperature, throughput, cooling cost, and inference accuracy in NPU chips. Our work unveils that hybrid thermal management, which composes different means to reduce the NPU temperature, is a key. To achieve that, we propose and implement PFS-TE technique that couples precision and FS together with superlattice TE cooling for effective NPU thermal management. Using commercial signoff tools, we obtain accurate power and timing analysis of MAC arrays after a full-chip design is performed based on 14-nm Intel FinFET technology. Then, multiphysics simulations using finite-element methods are carried out for accurate heat simulations in the presence and absence of on-chip cooling. Afterward, comprehensive design-space exploration is presented to demonstrate the Pareto frontier and the existing tradeoffs between temperature reductions, power overheads due to cooling, throughput, and inference accuracy. Using a wide range of NNs trained for image classification, experimental results demonstrate that our novel NPU thermal management increases the inference efficiency (TOPS/Joule) by $1.33\times $ , $1.87\times $ , and $2\times $ under different temperature constraints; 105 °C, 85 °C, and 70 °C, respectively, while the average accuracy drops merely from 89.0% to 85.5%. [ABSTRACT FROM AUTHOR]

Leveraging the inherent error resilience of a large number of application domains, approximate computing is established as an efficient design alternative to improve their energy profile. In this brief, we design energy optimal cross-layer approximate arithmetic circuits by enabling the efficient application of voltage overscaling (VOS). Departing from the conventional approaches followed today, we introduce the voltage-driven functional approximation and present the VoltAge-Driven nEtlist pRuning (VADER) framework. VADER is an automated synthesis framework that can be seamlessly integrated in any hardware design flow and implements a voltage-driven gate-level netlist pruning. Experimental evaluation shows that VADER reduces the error of the VOS application by 52% on average and delivers on average designs with 34% higher energy savings compared to state-of-the-art approximate adders and multipliers. [ABSTRACT FROM AUTHOR]

Approximate computing forms a promising paradigm shift for energy efficient design by aggressively decreasing power consumption of inherently error-tolerant applications. However, approximate computing architectures exacerbate the design complexity due to the diversity of inexact techniques and their impact on final circuit implementations. In this brief, we introduce approximate accelerator synthesis, which enables the design of power-optimized inexact circuits under error bound constraints, by leveraging the incorporation of diverse multi-level approximate techniques. We show the high efficiency of adopting multi-level approximation techniques and present a systematic approach for integrating multi-level approximation in hardware accelerator synthesis under error bound and voltage island constraints. [ABSTRACT FROM AUTHOR]

In this article, we target approximate computing for arithmetic circuits, focusing on the most complex and power-hungry units: hardware multipliers. Driven by the lack of a clear solution on the energy-error efficiency of existing approximate multiplication techniques, we present a new, efficient, and easily applied approximation design, as well as explore the current state-of-the-art design space. We show that the proposed approximation scheme can be equally applied at design time to enable synthesis of customized approximate multiplier circuits and at runtime to support dynamic approximation tuning scenarios. We achieve significant gains-up to 69-percent energy and 64-percent area savings with respect to accurate designs-by proposing hybrid approximation performed by two independent techniques that reduce both the depth (through perforation) and the width (through rounding) of the partial products accumulation tree. The corresponding runtime approximation solution delivers energy gains of up to 47 percent, introducing negligible area. More importantly, we show that design solutions configured through the proposed approach form the Pareto frontier of the energy-error space when considering direct quantitative comparisons with existing state of the art. [ABSTRACT FROM AUTHOR]

Approximate computing emerges as a new design paradigm for generating energy-efficient computing systems. Voltage overscaling (VOS) forms a very promising technique to generate approximate circuits, and its application in cooperation to other approximate techniques is proven to lead to more efficient solutions. However, the existing design tools fail to provide effective voltage-aware simulation for early exploration of power-error approximate design tradeoffs. In this brief, we propose VOSsim, a framework that extends state-of-the-art industry strength tools, to enable fast and accurate simulations of voltage overscaled circuits. We extensively evaluate VOSsim showing that it attains 99.2% output and 98.4% power accuracy, with an average speedup of $32\times $ in simulation time compared to high-precision SPICE simulations, i.e., the only available solution today for VOS-aware simulation. [ABSTRACT FROM AUTHOR]

Approximate computing forms a design alternative that exploits the intrinsic error resilience of various applications and produces energy-efficient circuits with small accuracy loss. In this paper, we propose an approximate hybrid high radix encoding for generating the partial products in signed multiplications that encodes the most significant bits with the accurate radix-4 encoding and the least significant bits with an approximate higher radix encoding. The approximations are performed by rounding the high radix values to their nearest power of two. The proposed technique can be configured to achieve the desired energy-accuracy tradeoffs. Compared with the accurate radix-4 multiplier, the proposed multipliers deliver up to 56% energy and 55% area savings, when operating at the same frequency, while the imposed error is bounded by a Gaussian distribution with near-zero average. Moreover, the proposed multipliers are compared with state-of-the-art inexact multipliers, outperforming them by up to 40% in energy consumption, for similar error values. Finally, we demonstrate the scalability of our technique. [ABSTRACT FROM AUTHOR]

Approximate computing has received significant attention as a promising strategy to decrease power consumption of inherently error tolerant applications. In this paper, we focus on hardware-level approximation by introducing the partial product perforation technique for designing approximate multiplication circuits. We prove in a mathematically rigorous manner that in partial product perforation, the imposed errors are bounded and predictable, depending only on the input distribution. Through extensive experimental evaluation, we apply the partial product perforation method on different multiplier architectures and expose the optimal architecture–perforation configuration pairs for different error constraints. We show that, compared with the respective exact design, the partial product perforation delivers reductions of up to 50% in power consumption, 45% in area, and 35% in critical delay. In addition, the product perforation method is compared with the state-of-the-art approximation techniques, i.e., truncation, voltage overscaling, and logic approximation, showing that it outperforms them in terms of power dissipation and error. [ABSTRACT FROM PUBLISHER]

Hardware acceleration has been proved an extremely promising implementation strategy for the digital signal processing (DSP) domain. Rather than adopting a monolithic application-specific integrated circuit design approach, in this brief, we present a novel accelerator architecture comprising flexible computational units that support the execution of a large set of operation templates found in DSP kernels. We differentiate from previous works on flexible accelerators by enabling computations to be aggressively performed with carry-save (CS) formatted data. Advanced arithmetic design concepts, i.e., recoding techniques, are utilized enabling CS optimizations to be performed in a larger scope than in previous approaches. Extensive experimental evaluations show that the proposed accelerator architecture delivers average gains of up to 61.91% in area-delay product and 54.43% in energy consumption compared with the state-of-art flexible datapaths. [ABSTRACT FROM PUBLISHER]