CloudMatrix 384 super-node uses 6912 x 400G OSFP silicon photonic (SiPh) Linear Drive Pluggable Optics (LPO) optical modules and 3168 fibers, connecting 384 Ascend 910C computing chips through a complete mesh interconnection architecture. Deep dive into Huawei's AI super-node: 384 Ascend 910C chips, 6912×400G OSFP SiPh LPO modules (1:18 ratio), 1. 2% failures stem from optics & how QSFPTEK cuts costs by 69. On May 14, 2025, the "2025 Chip and Optical Forum" hosted by HiSilicon and organized by. To address this, Macom and NVIDIA first proposed Linear-drive Pluggable Optics (LPO) in 2022. Its core concept is to remove digital processing units such as DSPs and CDRs from the module, constructing a purely analog "linear direct-drive" optical link. In the LPO architecture: The transmitter uses. The global AI computing industry has introduced supernode technology to address this issue. Large AI models involve both scale-up and scale-out networks, and AI computing supernode technology aims to integrate hundreds of GPUs/NPUs to form high-density computing units, such as the NVIDIA GB200. Eyeing so-called SuperNodes – singular supercomputing clusters with upwards of a million chips – Huawei is showing off its United Bus mesh (UB-mesh) technology., May 4, 2026 – GlobalFoundries (Nasdaq: GFS) (GF) today announced the introduction of its SCALE™ optical module solution for co-packaged optics (CPO). At the Apsara Conference 2025, Alibaba Cloud unveiled its new-generation Panjiu AI Infra 2.