Software-defined optical networking applications enabled by programmable integrated photonics
[EN] Data center networks are experiencing unprecedented exponential growth, mostly driven by the continuous computing demands in machine learning and artificial intelligence algorithms. Within this realm, optical networking offers numerous advantages, including low latency, energy efficiency, and b...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universitat Politècnica de València (UPV) |
| Repositorio: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia |
| Idioma: | inglés |
| OAI Identifier: | oai:riunet.upv.es:10251/228809 |
| Acceso en línea: | https://riunet.upv.es/handle/10251/228809 |
| Access Level: | acceso abierto |
| Palabra clave: | Circuit switching Integrated photonics Multiple input multiple output Optical circuits Optical interconnects Optical networks |
| Sumario: | [EN] Data center networks are experiencing unprecedented exponential growth, mostly driven by the continuous computing demands in machine learning and artificial intelligence algorithms. Within this realm, optical networking offers numerous advantages, including low latency, energy efficiency, and bandwidth transparency, positioning it as a compelling alternative to its electronic counterparts. In this work, we showcase a range of software-defined optical networking applications deployed on a general-purpose programmable integrated photonic processor. Leveraging graph-based theory, we experimentally demonstrate dynamic optical interconnects, circuit switching, and multicasting on the same photonic platform, yielding remarkable results in terms of crosstalk and reconfiguration speed. Our approach harnesses the benefits of reconfigurability and reliability, paving the way for a new generation of high-performance optical devices tailored for data center and computing clusters. |
|---|