At the 2026 Open AI Infra Summit, Alibaba Cloud showcased the evolution of the UALink 2.0 protocol and its integration with CXL, detailing new specifications, in‑network compute capabilities, and ecosystem developments that aim to overcome scale‑up bottlenecks in AI training and inference.
UALink is an open, high‑performance interconnect standard designed to link thousands of AI accelerators, offering NVLink‑level bandwidth, low latency, scalability, cost efficiency, and flexible topologies to meet the demanding communication needs of modern AI workloads.
This article examines the differences between Scale‑Out and Scale‑Up networking in AI compute clusters, comparing PCIe, Ethernet, InfiniBand, NVLink, UALink, and emerging standards like UB‑Mesh, and explains how each technology impacts bandwidth, latency, scalability, and cost for large‑scale model training.
The UALink 1.0 specification, driven by AMD, Intel, Broadcom and other industry leaders, introduces an open, low‑latency, high‑bandwidth interconnect that can link up to 1,024 AI accelerators, offering a cost‑effective alternative to Nvidia’s NVLink and reshaping the AI‑HPC market.
At the CCF China Storage Conference 2024 in Guangzhou, Alibaba Cloud’s research team presented the ALS (ALink System) ecosystem, discussed Scale‑Up interconnect protocols, and demonstrated multi‑layer storage innovations for AI workloads, highlighting hardware‑software integration, high‑bandwidth networking, and future CXL/PIM research.
The article analyzes the growing demand for high‑bandwidth, low‑latency GPU Scale‑Up interconnects in AI clusters, explains why existing Ethernet and RDMA solutions fall short, and examines the industry‑wide UALink alliance and Alibaba's Alink System as a new open‑ecosystem solution.
