Storage Reliability of Multi-bit Flash Oriented to Deep Neural Network

Xiang, Yachen; Kang, Jinfeng; Huang, Peng; Yang, Haozhang; Wang, K. L.; Han, Runze; Shen, Wenwen; Feng, Yang; Liu, C.; Liu, X. Y.

doi:10.1109/iedm19573.2019.8993508

Cited by 14 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the MNIST recognition task, the classification of one pattern takes < 1 μs time and ~20 nJ energy -both numbers > 10 3 × better than those of the 28-nm IBM TrueNorth digital chip for the same task at a similar fidelity. Xiang et al also have made an effort at NOR flash-based neuromorphic computing to eliminate the additional analog-to-digital/digital-to-analog (AD/DA) conversion, improve the reliability of multi-bit storage [122,123] . Compared to memristors, flash memory gains much fewer benefits on the cell size, operation voltage, and program/erase endurance although the mature fabrication process and suffers from the same scaling dilemma as a traditional transistor does.…”

Section: Mac Operation In Other Nonvolatile Devicesmentioning

confidence: 99%

Multiply accumulate operations in memristor crossbar arrays for analog computing

Chen

et al. 2021

J. Semicond.

View full text Add to dashboard Cite

Memristors are now becoming a prominent candidate to serve as the building blocks of non-von Neumann in-memory computing architectures. By mapping analog numerical matrices into memristor crossbar arrays, efficient multiply accumulate operations can be performed in a massively parallel fashion using the physics mechanisms of Ohm’s law and Kirchhoff’s law. In this brief review, we present the recent progress in two niche applications: neural network accelerators and numerical computing units, mainly focusing on the advances in hardware demonstrations. The former one is regarded as soft computing since it can tolerant some degree of the device and array imperfections. The acceleration of multiple layer perceptrons, convolutional neural networks, generative adversarial networks, and long short-term memory neural networks are described. The latter one is hard computing because the solving of numerical problems requires high-precision devices. Several breakthroughs in memristive equation solvers with improved computation accuracies are highlighted. Besides, other nonvolatile devices with the capability of analog computing are also briefly introduced. Finally, we conclude the review with discussions on the challenges and opportunities for future research toward realizing memristive analog computing machines.

show abstract

Section: Mac Operation In Other Nonvolatile Devicesmentioning

confidence: 99%

Multiply accumulate operations in memristor crossbar arrays for analog computing

Chen

et al. 2021

J. Semicond.

View full text Add to dashboard Cite

show abstract

“…The higher degree of freedom of computation puts higher requirements on the memory device, meaning that the device unit should store more states, which is different from the binary switching characteristics. At present, device technologies that can implement multiple state storage include RRAM, [45] PCM, [52] flash, [147] and so on. Usually, a suitable electrical signal such as an accumulated pulse is applied to these memory devices, and the resistance of the device evolves accordingly.…”

Section: Hybrid Precision Digital Computingmentioning

confidence: 99%

In-memory computing to break the memory wall*

Huang¹,

Liu²,

Jiang³

et al. 2020

Chinese Phys. B

View full text Add to dashboard Cite

Facing the computing demands of Internet of things (IoT) and artificial intelligence (AI), the cost induced by moving the data between the central processing unit (CPU) and memory is the key problem and a chip featured with flexible structural unit, ultra-low power consumption, and huge parallelism will be needed. In-memory computing, a non-von Neumann architecture fusing memory units and computing units, can eliminate the data transfer time and energy consumption while performing massive parallel computations. Prototype in-memory computing schemes modified from different memory technologies have shown orders of magnitude improvement in computing efficiency, making it be regarded as the ultimate computing paradigm. Here we review the state-of-the-art memory device technologies potential for in-memory computing, summarize their versatile applications in neural network, stochastic generation, and hybrid precision digital computing, with promising solutions for unprecedented computing tasks, and also discuss the challenges of stability and integration for general in-memory computing.

show abstract

“…[14][15][16][17] However, at present, most of the studies concentrate on the prediction model of retention characteristics and error rate of Flash. [18,19] There are few prediction models and corresponding solutions for the endurance and retention of RRAM chips. Iwasaki.…”

Section: Introductionmentioning

confidence: 99%

Lifetime Improvement of 28 nm Resistive Random Access Memory Chip by Machine Learning‐Assisted Prediction Model Collaborated with Resurrection Algorithm

Zheng,

Wu,

Xie

et al. 2024

Adv Elect Materials

View full text Add to dashboard Cite

In this work, a machine learning‐assisted prediction model is proposed to analyze the reliability issues in the 28 nm resistive random access memory (RRAM) chip with raw data measured from RRAM test chip. The neural network of long‐short time memory (LSTM) is trained by the voltages and resistance during the endurance test (input vectors) and generates the output of the dichotomy states with a satisfied testing result >83.08%. According to the prediction results, the “real fail” state or “fake fail” state of the devices can get in the future. By collaborating with a well‐designed resurrection algorithm (RA), the percentage of real and fake failed cells dropped by 35% and 29%, respectively. Besides, the tail bits in retention significantly reduced from 33% to 14.6% due to the reduction of oxygen vacancies in the gaps of conducting filaments by applying ten consecutive cycles of RA. This intelligent prediction and repair module can prolong the lifetime of RRAM chips effectively in practical applications.

show abstract

Storage Reliability of Multi-bit Flash Oriented to Deep Neural Network

Cited by 14 publications

References 7 publications

Multiply accumulate operations in memristor crossbar arrays for analog computing

Multiply accumulate operations in memristor crossbar arrays for analog computing

In-memory computing to break the memory wall*

Lifetime Improvement of 28 nm Resistive Random Access Memory Chip by Machine Learning‐Assisted Prediction Model Collaborated with Resurrection Algorithm

Contact Info

Product

Resources

About