Optical in-memory computing using laser array

Omar Alkhazragi

doi:10.1038/s41377-026-02338-x

Optical in-memory computing using laser array

Omar Alkhazragi^*

^*Corresponding author for this work

Department of Electrical Engineering

Research output: Contribution to journal › Comment/debate

Abstract

A new optical in-memory computing system based on an array of vertical-cavity surface-emitting lasers (VCSELs) has the potential to circumvent the Von Neumann bottleneck. The high modulation speed of the lasers in the array allows for fast computing and their high efficiency can enable edge computing in autonomous vehicles and drones. This efficient, highly scalable system was demonstrated to perform 900 million convolutions per second with 98% computing accuracy.

Original language	English
Article number	262
Journal	Light: Science and Applications
Volume	15
Issue number	1
DOIs	https://doi.org/10.1038/s41377-026-02338-x
State	Published - Dec 2026

Bibliographical note

Publisher Copyright:
© The Author(s) 2026.

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Atomic and Molecular Physics, and Optics

Access to Document

10.1038/s41377-026-02338-x

Cite this

@article{18dd79c37e0e45bebdde7f73855340e8,

title = "Optical in-memory computing using laser array",

abstract = "A new optical in-memory computing system based on an array of vertical-cavity surface-emitting lasers (VCSELs) has the potential to circumvent the Von Neumann bottleneck. The high modulation speed of the lasers in the array allows for fast computing and their high efficiency can enable edge computing in autonomous vehicles and drones. This efficient, highly scalable system was demonstrated to perform 900 million convolutions per second with 98\% computing accuracy.",

author = "Omar Alkhazragi",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2026.",

year = "2026",

month = dec,

doi = "10.1038/s41377-026-02338-x",

language = "English",

volume = "15",

journal = "Light: Science and Applications",

issn = "2047-7538",

publisher = "Springer Nature",

number = "1",

}

TY - JOUR

T1 - Optical in-memory computing using laser array

AU - Alkhazragi, Omar

PY - 2026/12

Y1 - 2026/12

N2 - A new optical in-memory computing system based on an array of vertical-cavity surface-emitting lasers (VCSELs) has the potential to circumvent the Von Neumann bottleneck. The high modulation speed of the lasers in the array allows for fast computing and their high efficiency can enable edge computing in autonomous vehicles and drones. This efficient, highly scalable system was demonstrated to perform 900 million convolutions per second with 98% computing accuracy.

AB - A new optical in-memory computing system based on an array of vertical-cavity surface-emitting lasers (VCSELs) has the potential to circumvent the Von Neumann bottleneck. The high modulation speed of the lasers in the array allows for fast computing and their high efficiency can enable edge computing in autonomous vehicles and drones. This efficient, highly scalable system was demonstrated to perform 900 million convolutions per second with 98% computing accuracy.

UR - https://www.scopus.com/pages/publications/105040759149

U2 - 10.1038/s41377-026-02338-x

DO - 10.1038/s41377-026-02338-x

M3 - Comment/debate

AN - SCOPUS:105040759149

SN - 2047-7538

VL - 15

JO - Light: Science and Applications

JF - Light: Science and Applications

IS - 1

M1 - 262

ER -

Optical in-memory computing using laser array

Abstract

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this