Nonvolatile Processor Architectures: Efficient, Reliable Progress with Unstable Power

Ma, Kaisheng; Li, Xueqing; Swaminathan, Karthik; Yang, Zheng; Li, Shuangchen; Liu, Yongpan; Xie, Yuan; Sampson, John J.; Narayanan, Vijaykrishnan

doi:10.1109/mm.2016.35

Cited by 38 publications

(20 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sanyam et al in their recent research [93] on wear bench application showed that high performance can be achieved for deep learning along with other wearable applications (e.g., image compression, audio play back and video rendering) using Out of Order 32 bit CPU core with SIMD feature (e.g., ARM-A15 with NEON). Non-volatile processor architecture has been proposed [94] for energy harvesting IoT device to maximize the forward progress of IoT application in unstable and intermittent power supply environment. This paper considers out-of-order core as a base line processor for IoT wearable applications and introduces memristive accelerator to offload compute intensive CNN calculations from the core not only to reduce core dynamic power consumption but also to limit data movement between memory and core.…”

Section: Mobile Iot Applicationsmentioning

confidence: 99%

MB-CNN: Memristive Binary Convolutional Neural Networks for Embedded Mobile Devices

Chowdhury

Kulkarni

Bojnordi

2018

JLPEA

View full text Add to dashboard Cite

Applications of neural networks have gained significant importance in embedded mobile devices and Internet of Things (IoT) nodes. In particular, convolutional neural networks have emerged as one of the most powerful techniques in computer vision, speech recognition, and AI applications that can improve the mobile user experience. However, satisfying all power and performance requirements of such low power devices is a significant challenge. Recent work has shown that binarizing a neural network can significantly improve the memory requirements of mobile devices at the cost of minor loss in accuracy. This paper proposes MB-CNN, a memristive accelerator for binary convolutional neural networks that perform XNOR convolution in-situ novel 2R memristive data blocks to improve power, performance, and memory requirements of embedded mobile devices. The proposed accelerator achieves at least 13.26 × , 5.91 × , and 3.18 × improvements in the system energy efficiency (computed by energy × delay) over the state-of-the-art software, GPU, and PIM architectures, respectively. The solution architecture which integrates CPU, GPU and MB-CNN outperforms every other configuration in terms of system energy and execution time.

show abstract

Section: Mobile Iot Applicationsmentioning

confidence: 99%

MB-CNN: Memristive Binary Convolutional Neural Networks for Embedded Mobile Devices

Chowdhury

Kulkarni

Bojnordi

2018

JLPEA

View full text Add to dashboard Cite

show abstract

“…Broadly, these can be classified into two categories; software-based approaches [28,29], where the system's state (e.g. core and general-purpose registers, and main RAM memory) is saved into a Non-Volatile Memory (NVM) before a power failure occurs and restored once the power supply recovers, and hardware-assisted approaches [30,31], where the entire system is designed to be non-volatile (e.g. a unified low-power NVM system with non-volatile core and general-purpose registers).…”

Section: Transient Computing Techniquesmentioning

confidence: 99%

Momentum

Balsamo

Fletcher

Weddell

et al. 2018

ACM Trans. Embed. Comput. Syst.

View full text Add to dashboard Cite

Recent research has looked to supplement or even replace the batteries in embedded computing systems with energy harvesting, where energy is derived from the device's environment. However, such supplies are generally unpredictable and highly variable, and hence systems typically incorporate large external energy buffers (e.g. supercapacitors) to sustain computation; however, these pose environmental issues and increase system size and cost. This paper proposes Momentum, a general power-neutral methodology, with intrinsic system-wide maximum power point tracking, that can be applied to a wide range of different computing systems, where the system dynamically scales its performance (and hence power consumption) to optimize computational progress depending on the power availability. Momentum enables the system to operate around an efficient operating voltage, maximizing forward application execution, without adding any external tracking or control units. This methodology combines at run-time 1) a hierarchical control strategy which utilizes available power management controls (such as dynamic voltage and frequency scaling, and core hot-plugging) to achieve efficient power-neutral operation, 2) a software-based maximum power point tracking scheme (unlike existing approaches, this does not require any additional hardware), which adapts the system power consumption so that it can work at the optimal operating voltage, considering the efficiency of the entire system rather than just the energy harvester, and 3) experimental validation on two different scales of computing system: a low power microcontroller (operating from the already-present 4.7 µF decoupling capacitance) and a multi-processor system-on-chip (operating from 15.4 mF added capacitance). Experimental results from both a controlled supply and energy harvesting source show that Momentum operates correctly on both platforms, and exhibits improvements in forward application execution of up to 11% when compared to existing power-neutral approaches, and 46% compared to existing static approaches.

show abstract

“…Architectural approaches (sometimes referred to as NonVolatile Processors, or NVPs) have also been proposed for transient computing, which provide hardware support for maintaining and saving state using NVM elements [10].…”

Section: B Transient Computingmentioning

confidence: 99%

Energy-driven computing: Rethinking the design of energy harvesting systems

Merrett

Al-Hashimi

2017

Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE), 2017

View full text Add to dashboard Cite

Abstract-Energy harvesting computing has been gaining increasing traction over the past decade, fueled by technological developments and rising demand for autonomous and battery-free systems. Energy harvesting introduces numerous challenges to embedded systems but, arguably the greatest, is the required transition from an energy source that typically provides virtually unlimited power for a reasonable period of time until it becomes exhausted, to a power source that is highly unpredictable and dynamic (both spatially and temporally, and with a range spanning many orders of magnitude). The typical approach to overcome this is the addition of intermediate energy storage/buffering to smooth out the temporal dynamics of both power supply and consumption. This has the advantage that, if correctly sized, the system 'looks like' a battery-powered system; however, it also adds volume, mass, cost and complexity and, if not sized correctly, unreliability. In this paper, we consider energydriven computing, where systems are designed from the outset to operate from an energy harvesting source. Such systems typically contain little or no additional energy storage (instead relying on tiny parasitic and decoupling capacitance), alleviating the aforementioned issues. Examples of energy-driven computing include transient systems (which power down when the supply disappears and efficiently continue execution when it returns) and power-neutral systems (which operate directly from the instantaneous power harvested, gracefully modulating their consumption and performance to match the supply). In this paper, we introduce a taxonomy of energy-driven computing, articulating how power-neutral, transient, and energy-driven systems present a different class of computing to conventional approaches.

show abstract

Nonvolatile Processor Architectures: Efficient, Reliable Progress with Unstable Power

Cited by 38 publications

References 7 publications

MB-CNN: Memristive Binary Convolutional Neural Networks for Embedded Mobile Devices

MB-CNN: Memristive Binary Convolutional Neural Networks for Embedded Mobile Devices

Momentum

Energy-driven computing: Rethinking the design of energy harvesting systems

Contact Info

Product

Resources

About