Technology

Huawei unveils Atlas 950 SuperCluster — promises 1 ZettaFLOPS FP4 performance and features hundreds of thousands of 950DT APUs

Source

Tom's Hardware

Published

Sep 19, 2025

TL;DR

AI Generated

Huawei has introduced the Atlas 950 SuperCluster, a high-performance AI solution featuring 1 ZettaFLOPS FP4 performance and hundreds of thousands of Ascend 950DT NPUs. The system is designed to compete with Nvidia's Rubin-based systems by late 2026. The SuperCluster comprises 64 Atlas 950 SuperPoDs with 524,288 AI accelerators spread across over 10,240 interconnected cabinets, offering up to 524 FP8 ExaFLOPS for AI training and 1 FP4 ZettaFLOPS for AI inference. Huawei positions the SuperCluster to support large-scale AI models with billions to trillions of parameters, emphasizing compute throughput, interconnect bandwidth, and system stability. Huawei also revealed plans for the Atlas 960 SuperCluster in late 2027, promising even higher performance with over 1 million Ascend 960 NPUs.

Read Full Article

Panel-Level Packaging’s Second Wave Meets Engineering Reality

Panel-level packaging is gaining traction due to economic pressures and the increasing size of AI accelerators and HPC packages. Glass substrates are being explored to address warpage and dimensional stability issues, but they introduce new failure modes that require material solutions. Challenges in panel-level processing include materials and process integration, not just packaging problems. The industry is moving towards panels driven by economic and technological shifts, but solving these challenges requires a holistic approach.

SemiEngineering•

2 weeks ago

SemiEngineering

Inside the AI Accelerator: Essential IP Design Solutions: eBook

The eBook delves into how advanced IP, high-speed interconnects, memory interfaces, and multi-die architectures are utilized in next-gen AI accelerators to surpass single-chip limitations. It highlights the role of optical links in enhancing bandwidth and security IP in safeguarding AI data without compromising performance. The eBook also covers how technologies like UALink, PCIe, CXL, and Ultra Ethernet support scaling AI architectures, integrating compute, memory, and accelerators, and enhancing bandwidth density through optical I/O. The focus is on unlocking AI performance at scale and ensuring data security across accelerators.

SemiEngineering•

3 weeks ago

Industry's first TSMC COUPE-based optical connectivity solution for next-gen AI chips displayed — Alchip and Ayar Labs show future silicon photonics device

Alchip and Ayar Labs showcased a TSMC COUPE-based optical connectivity solution at TSMC's European OIP forum, designed for next-gen AI accelerators. This solution integrates Ayar's silicon-photonics TeraPHY IC with Alchip's electrical interface die and detachable fiber connector, offering up to 100 Tb/s of bandwidth per accelerator. The system targets hardware developers seeking optical connectivity without the need to build their own optical subsystem. The three-chiplet co-packaged optical I/O subsystem includes a protocol-converter chiplet, EIC, and TeraPHY PIC, enabling high-speed optical modulation and detection. This production-ready solution allows smaller chip designers to incorporate optical connectivity affordably and efficiently.

Tom's Hardware•

5 months ago

SiPearl unveils Europe's first dual-use sovereign processor with 80 cores — expected in 2027 for government, aerospace, and defense applications

SiPearl has unveiled Athena1, Europe's first dual-use processor with up to 80 cores, catering to government, aerospace, and defense applications. The CPU, based on Arm’s Neoverse V1 architecture, is set to be available in various core configurations by the second half of 2027. SiPearl's Athena1, an enhanced version of its Rhea processor, boasts advanced encryption and security features suitable for tasks like encrypted communications and intelligence analysis. The processor, manufactured by TSMC, aims to bolster Europe's technological sovereignty in the face of geopolitical challenges.