首页|A Comprehensive Survey on Integrated Photonic Switch Fabrics: From Device to Network Topology and Controller Design

A Comprehensive Survey on Integrated Photonic Switch Fabrics: From Device to Network Topology and Controller Design

扫码查看
Emerging applications such as Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data are significantly increasing data traffic within data center networks and high-performance computing (HPC) systems. This surge in data flow is pushing the existing network infrastructures to their limits. To address the demands of this data-driven era, there is a critical need for network technologies and architectures that offer higher capacity, improved performance, and reduced transmission delays. Central to these networks are switches, which play a crucial role in directing data traffic between various nodes and clusters with minimal latency. These switches must manage large bandwidths and support numerous ports efficiently. As electronic switches approach limitations in energy efficiency and bandwidth scaling, optical circuit switching has emerged as a promising alternative. Optical switches are being explored for their potential to provide higher bandwidth and lower energy consumption in data center and HPC networks. This survey paper reviews the most recent advancements in various types of optical switches, including those based on Micro-Electromechanical Systems (MEMS), Silicon Photonics, and III-V interferometric switches, as well as wavelength-selective and hybrid solutions. It covers aspects such as device design, switch fabric topology, tuning mechanisms (e.g., thermo-optic and electro-optic), and control plane implementations. The paper highlights the potential of these technologies while also discussing some challenges that remain in this field for further advancement of optical switches.

Optical switchesOptical interferometryTuningSwitchesBandwidthOptical devicesOptical fiber networksControl systemsSwitching circuitsHigh-speed optical techniques

Mohammad Amin Mahdian、Ebadollah Taheri、Mahdi Nikdast

展开 >

Department of Electrical and Computer Engineering, Colorado State University, Fort Collins, CO, USA

2025

IEEE Access

IEEE Access

ISSN:
年,卷(期):2025.13(1)
  • 300