Review of the application of fgOTN in the interconnection of artificial intelligence data centers

Abstract

With the rapid growth of data traffic and service requirements in artificial intelligence data centers, efficient and flexible network solutions have become critical. The fine grain optical transport network (fgOTN), which serves as a continuation of synchronous digital hierarchy (SDH) technology and an extension of optical transport network (OTN) technology, is applied in the interconnection of artificial intelligence data centers to meet their multiple requirements for flexible scheduling, efficient transmission, strict security isolation, and low latency. Firstly, the basic concepts, technical architecture, and application scenarios of fgOTN were introduced. Subsequently, the related concepts, architectures, key technologies, and application scenarios of artificial intelligence data centers were elaborated. Based on these foundations, the application of fgOTN in interconnection of artificial intelligence data centers was discussed in detail, aiming to promote efficient and reliable data transmission between artificial intelligence data centers. Finally, the research directions and development trends were discussed.

Keywords

• fgOTN
• interconnection of artificial intelligence data center
• application scenario