Siemon, a global leader in network infrastructure solutions, has introduced a new guide aimed at assisting data center professionals in making informed decisions regarding network infrastructure to support current and future AI advancements
The guide offers valuable insights for a wide range of users, from AI cloud service providers needing power and cooling solutions for high-density rack configurations in large high-performance computing (HPC) clusters, to large enterprises aiming to integrate AI infrastructure into existing on-site or colocation data centres.
Addressing the unique demands of data- and compute-intensive machine learning (ML) and generative AI applications, which require extremely high bandwidth and low latency, the guide recommends AI-ready fibre solutions. These include high-density, end-to-end singlemode and multimode MTP fibre systems that support ultra low-loss (ULL) transmission up to 800G and beyond for compute and storage fabrics. Additionally, the guide suggests considering direct attach cables (DACs) and active optical cables (AOCs) for high-speed, low-latency point-to-point connections within back-end AI clusters, catering to Ethernet, RoCE, and InfiniBand networks.
Siemon’s new guide also provides advice on suitable cabling configurations in order to meet the higher power and cooling demands of GPU-based servers and it discusses the latest developments surrounding Infiniband and Ethernet, suggesting connectivity options that support both technologies.
“As data centres gear up for AI, users need innovative, robust network infrastructure solutions that will help them to easily design, deploy, and scale back-end, front-end, and storage network fabrics for complex high-performance computing (HPC) AI environments,” says vice-president sales for India, the Middle East and Africa at Siemon. “We have created this guide to provide data centre professionals with a detailed overview of what they need for the design, delivery, and day-to-day operation of AI networks.”