UnifiedBus

Huawei introduced SuperPoD at HUAWEI CONNECT 2025, a new AI infrastructure architecture that aggregates thousands of AI chips into a unified resource using UnifiedBus (UB). This creates a “supercomputer” from distributed servers, designed to address the limitations of traditional architectures. The Atlas 950 SuperPoD utilizes up to 8,192 Ascend 950DT chips, with future plans for the larger Atlas 960. Beyond AI, TaiShan 950 SuperPoD targets general-purpose computing. Huawei’s open-source approach with UnifiedBus 2.0 aims to accelerate innovation and foster broad industry participation in AI infrastructure development.

At Huawei Connect 2025, Huawei revealed its Ascend chip roadmap, including the 950, 960, and 970 series for AI and HPC, challenging NVIDIA’s dominance. Despite semiconductor manufacturing challenges, Huawei focuses on domestic design, proprietary tech, and open-source strategies. New Ascend chips promise performance leaps with enhanced interconnects. Huawei’s SuperPoD and SuperCluster strategy, powered by UnifiedBus 2.0 (an open protocol), aims to provide scalable, high-performance computing, expanding into general-purpose computing with Kunpeng 950 processors and TaiShan SuperPod. Huawei claims significant performance advantages over competitors.