{"slug": "amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios", "title": "AMD Advancing AI 2026: Pensando \"Salina\" DPU and \"Vulcano\" AI NIC Power Helios Networking", "summary": "AMD detailed its Pensando \"Salina\" DPU and \"Vulcano\" AI NIC as key components of the Helios rack-scale networking strategy, designed to interconnect GPUs for AI models exceeding 10 trillion parameters. The company noted that bandwidth demands are doubling in under two years due to exponential growth in AI model sizes, from 65-million-parameter Transformers in 2017 to trillion-parameter systems today. AMD's solution includes a third-generation Salina DPU, a first-generation UALink-over-Ethernet scale-up fabric, and a second-generation Vulcano AI NIC, all managed by Helios software.", "body_md": "Networking is the third pillar of AMD's Advancing AI 2026 rack-scale strategy, and today the company detailed the AMD Pensando silicon and fabric technology that connects the Helios rack. As Helios is AMD's first rack-scale design, the company encountered challenges in interconnecting many GPUs, memories, and data movement speeds within a single system, both for scaling up and scaling out for multi-rack connections inside the data center. Current AI infrastructure must closely follow model development as AI model sizes have grown exponentially, from the original 65-million-parameter Transformer in 2017 to trillion-parameter GPT-4-class systems and now 10-trillion-plus parameter agentic and reasoning models. To run a model with more than 10 trillion parameters, you need to connect dozens of GPUs with high-speed VRAM, as well as a high-speed interconnect to split model weights effectively so latency isn't an issue. The modern interconnect of choice is again the plain-old Ethernet, not proprietary interconnects, which have become the standard for AI networking.\n\nAMD's solution spans three layers: a third-generation Pensando \"Salina\" DPU on the front end, a first-generation UALink-over-Ethernet (UALoE) scale-up fabric inside the rack, and a second-generation Pensando \"Vulcano\" AI NIC for scale-out. All of this is managed by AMD's Helios management software and Fabric Manager. AMD outlined challenges that bandwidth needs are doubling in less than two years driven by massive AI models growing at a rapid pace, and that the hardware iteration cycles need to account for intelligence in the data fabric layer, not only compute layer.", "url": "https://wpnews.pro/news/amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios", "canonical_source": "https://www.techpowerup.com/351007/amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios-networking", "published_at": "2026-07-23 19:30:44+00:00", "updated_at": "2026-07-23 20:00:24.872124+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-infrastructure", "ai-chips"], "entities": ["AMD", "Pensando", "Salina", "Vulcano", "Helios", "UALink"], "alternates": {"html": "https://wpnews.pro/news/amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios", "markdown": "https://wpnews.pro/news/amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios.md", "text": "https://wpnews.pro/news/amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios.txt", "jsonld": "https://wpnews.pro/news/amd-advancing-ai-2026-pensando-salina-dpu-and-vulcano-ai-nic-power-helios.jsonld"}}