From Host Node To Heterogeneous Rack: Rethinking The AI CPU


AI infrastructure is entering a crucial new phase. The first phase of generative AI infrastructure was defined by accelerator scale: how many GPUs, NPUs or custom AI accelerators could be deployed, powered, cooled and connected. That phase is not over, but it is no longer sufficient. The next phase is about rack-scale system composition: heterogeneous AI racks where different compute resourc... » read more

Building A Production-Ready Optically Connected Rack For AI Scale-Up


By Nandita Aggarwal and Nicholas Chang As AI models drive compute demand, servers keep getting bigger. Rack‑scale AI systems (such as the 72-GPU systems from NVIDIA or AMD) enable many GPUs to work together through system-level optimization. They push beyond the limits of single-chip performance and meet the soaring compute needs of the AI era. But this is just the beginning. The next s... » read more