The WAIC 2026 showcase introduced eight domestic AI supernodes—from Huawei’s 1024‑card Atlas 950 to New H3C’s scalable UniPoD S80000—detailing their card counts, interconnect bandwidth, FP8/FP4 performance, and how China’s AI compute evaluation is shifting from single‑chip peaks to system‑level efficiency and cost.
The article provides a technical breakdown of Huawei's UnifiedBus (Lingqu) interconnect protocol for AI supernodes, detailing its full‑stack design that unifies chip‑, board‑, and cabinet‑level communication, its three‑layer stack (PHY, Transaction, Service), resource‑scheduling and virtualization capabilities, and its deployment in Atlas950/960 SuperPod clusters.
On May 30, 2026 Huawei released the Lingqu 2.0.1 specification, tightening protocol details, expanding heterogeneous compatibility, adding firmware and OS integration guides, and optimizing large‑scale routing to make the bus‑level, unified‑protocol, full‑pooling architecture more stable and ready for massive AI supernode deployments, while positioning it as an open alternative to Nvidia’s NVLink ecosystem.
The article explains what AI supernodes are, how they differ from traditional server clusters, and why their bus‑level interconnect, global memory pooling, peer‑to‑peer compute and integrated liquid‑cooled racks deliver up to 15× bandwidth gains, 4× inference concurrency, and significant cost reductions, while comparing the approaches of Nvidia, Huawei and other Chinese vendors and outlining future scaling challenges.
The article explains the concept of SuperNode (SuperPod) as a new AI‑focused compute infrastructure, outlines its high‑density integration, ultra‑fast interconnects, and unified resource management, and compares three leading implementations from NVIDIA, Huawei, and the ETH‑X project.
Huawei’s 2025 Full‑Connection Conference unveiled the Atlas 950 and Atlas 960 SuperPoD supernodes, detailing their massive card counts, unprecedented compute, memory and bandwidth capabilities, and explaining how their full‑stack hardware‑software design dramatically accelerates large‑model AI training and inference.
