Photonic devices compute in memory

Varnava, Christiana

doi:10.1038/s41928-019-0226-1

Cited by 4 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Certainly, demand for machine intelligence and AI has skyrocketed and is continuing to do so for the foreseeable future. All this information processing demands more memory performance, especially in emerging non-van Neumann systems such as neural networks or tensor-core processors [1][2][3] .…”

Section: Introductionmentioning

confidence: 99%

Electrical Programmable Low-loss high cyclable Nonvolatile Photonic Random-Access Memory

Meng

Peserico

et al. 2022

Preprint

View full text Add to dashboard Cite

The retention-of-state functionality provided by memories is fundamental to any Turing machine and neural network, hence is critical for any information system today. While emerging optical machine learning accelerators and photonic neuromorphic computing paradigms provide promising signal processing and computing performance, the lack of a photon-photon force in the universe makes storing optical information challenging. Fortunately, phase change materials provide such a missing memristive nonvolatile function via their reconfigurable crystalline structure and allow for rapid optical READ paradigms. However, demonstrations of photonic memory are limited by high optical loss, low state-cyclability, and rely on cumbersome non-CMOS like optical programmability. To overcome all three shortcomings and unlock the full potential of optical information storage and access, here we introduce a photonic random-access memory featuring vanishing low optical loss, demonstrate more than half a million switching cycles, a 100x improvement over state-of-art, and realize electrical programmability on-chip. The exceedingly low optical absorption (0.0015 dB/μm) is achieved via a novel broadband transparent phase change material, Ge2Sb2Se5 integrated atop a nanophotonic waveguide of a silicon chip. We show a highly efficient signal modulation (0.2 dB/μm) achieved by realizing a newly designed paired micro-heaters along both sides of waveguide, which allows for electronic-standard programmability of these photonic memories. When interrogated by an optical beam, they offer picosecond-short memory READ latency. Furthermore, we demonstrate a partial amorphization scheme realizing multi-state memory levels on a single heater enhancing footprint efficiency. Lastly, we verify the energy and switching speed and show how each trades-off with heater-to-waveguide proximity and signal strength, respectively. Such as CMOS-near electronically programmed and optical read photonic random-access memory with low-optical loss yet efficient programmability can become a crucial building block for network edge AI system of the looming industry 4.0 era.

show abstract

Section: Introductionmentioning

confidence: 99%

Electrical Programmable Low-loss high cyclable Nonvolatile Photonic Random-Access Memory

Meng

Peserico

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Thus, researchers are seeking alternatives to the conventional von Neumann architecture, in which computations can be performed using computing units that are not physically separate from memory units; such technology is termed in-memory computing. [22][23][24][25][26][27][28][29][30][31] However, technologies that are more tangible and plausible and capable of surpassing the computing performance of the conventional von Neumann architecture are also available; these technologies can serve as a platform for increasing the information density per given unit device using ternary, quaternary, or even higher multivalued logic (MVL) systems. [15][16][32][33][34] Current processing systems are based on the binary system, wherein information is stored as a "0" bit or a "1" bit; in contrast, the MVL system functions as a ternary, quaternary, or even higher-valued system, which enables significant reductions in the number of devices and the overall system complexity.…”

Section: Introductionmentioning

confidence: 99%

Recent Advances on Multivalued Logic Gates: A Materials Perspective

Kang

Cho

2021

Advanced Science

View full text Add to dashboard Cite

The recent advancements in multivalued logic gates represent a rapid paradigm shift in semiconductor technology toward a new era of hyper Moore's law. Particularly, the significant evolution of materials is guiding multivalued logic systems toward a breakthrough gradually, whereby they are transcending the limits of conventional binary logic systems in terms of all the essential figures of merit, i.e., power dissipation, operating speed, circuit complexity, and, of course, the level of the integration. In this review, recent advances in the field of multivalued logic gates based on emerging materials to provide a comprehensive guideline for possible future research directions are reviewed. First, an overview of the design criteria and figures of merit for multivalued logic gates is presented, and then advancements in various emerging nanostructured materials-ranging from 0D quantum dots to multidimensional heterostructures-are summarized and these materials in terms of device design criteria are assessed. The current technological challenges and prospects of multivalued logic devices are also addressed and major research trends are elucidated.

show abstract

“…[ 5 ] The scaling down of MOSFET has been challenging as the power management of IC strongly influences the production price and reliability. Currently, various breakthroughs have been suggested to resolve the limitation in scaling down conventional MOSFETs, such as increasing the data density in a unit area (such as multilevel logic, [ 6–11 ] memory computing, [ 12–15 ] and other area‐efficient method [ 16–18 ] ) and improving in device architecture (such as FinFET, [ 19–22 ] gate‐all‐around (GAA) FET, [ 22–28 ] tunneling field‐effect transistor (TFET) [ 22,29–31 ] ).…”

Section: Introductionmentioning

confidence: 99%

“…[5] The scaling down of MOSFET has been challenging as the power management of IC strongly influences the production price and reliability. Currently, various breakthroughs have been suggested to resolve the limitation in scaling down conventional MOSFETs, such as increasing the data density in a unit area (such as multilevel logic, [6][7][8][9][10][11] memory computing, [12][13][14][15] and other area-efficient method [16][17][18] ) and improving in device architecture (such as FinFET, [19][20][21][22] gateall-around (GAA) FET, [22][23][24][25][26][27][28] tunneling field-effect transistor (TFET) [22,[29][30][31] ). Band-to-band tunneling (BTBT), which is the main mechanism to operate TFET, is one of the promising alternative architectures for conventional MOSFET to solve the power consumption issue.…”

mentioning

confidence: 99%

Band‐to‐Band Tunneling Control by External Forces: A Key Principle and Applications

Woo

Kim

Yoo

2022

Adv Elect Materials

View full text Add to dashboard Cite

Band‐to‐band tunneling (BTBT) devices with superior subthreshold swing directly related to on/off switching speed and power consumption efficiency have emerged as a breakthrough of the limitation in conventional metal‐oxide‐semiconductor field‐effect transistors (MOSFETs). However, it is difficult to reach a higher level of electrical characteristics with only a combination of materials based on their intrinsic characteristics. External forces, such as electric fields, light, and temperature, can modulate the electron and hole concentration and control the tunneling probability and the electrical characteristics of heterostructure electronics. Recent articles employing external forces to improve the BTBT performance and demonstrate the mechanism of BTBT devices are summarized with five representative external forces. Moreover, the utility of the external force‐induced performance improvement of the BTBT device is also discussed by providing various applications.

show abstract

Photonic devices compute in memory

Cited by 4 publications

References 0 publications

Electrical Programmable Low-loss high cyclable Nonvolatile Photonic Random-Access Memory

Electrical Programmable Low-loss high cyclable Nonvolatile Photonic Random-Access Memory

Recent Advances on Multivalued Logic Gates: A Materials Perspective

Band‐to‐Band Tunneling Control by External Forces: A Key Principle and Applications

Contact Info

Product

Resources

About