Huawei introduces the Atlas AI computing platform

The Huawei Atlas AI computing platform is powered by Huawei's Ascend series AI processors. Atlas offers various product form factors, such as accelerator modules, cards, AI edge stations, and appliances, providing an all-scenario AI infrastructure solution for the device-edge-cloud. It applies to fields of applications such as smart city, carriers, finance, Internet, and electric power.

Integral to Huawei's full-stack, all-scenario AI solution, the Atlas AI computing platform unlocks supreme compute power to help customers embrace an AI-fueled future.

The Huawei Ascend AI processors ride on the advantages of the cutting-edge Da Vinci 3D Cube architecture, delivering ultra-high computing density and energy efficiency to enable Atlas with unrivaled AI compute power. The Atlas series products feature the following highlights:

Atlas 200 AI Accelerator Module: Packaged in a form factor half the size of a credit card, it supports 16-channel real-time HD video analytics and can be deployed on terminal devices such as cameras, robots, and drones.

Atlas 200 DK AI Developer Kit: It helps developers set up the development environment in 30 minutes, and integrates a graphical programming environment to bring new levels of development efficiency. Thanks to the full-scenario capability of the Ascend processor, programs can be developed on the Atlas 200 DK AI Developer Kit in a one-off manner and deployed in device-edge-cloud scenarios with zero code modification.

Atlas 300 AI Accelerator Card: It adopts a half-height half-length form factor and provides standard PCIe interfaces. It provides 64 TOPS of INT8 computing performance and supports 64-channel real-time HD video analytics with a 67 W power consumption. The Atlas 300 also boasts a 32 GB memory, which greatly improves the performance of video analytics and high-density inference.

Atlas 500 AI Edge Station: Providing powerful performance, it can perform real-time data processing at the edge. A single device can provide 16 TOPS of INT8 processing capability with an ultra-low power consumption of less than 1 kWh per day. Built for complex natural environments at the edge, the Atlas 500 runs stably in harsh deployment conditions of -40°C to +70°C.

During the launch, Hou Jinlong, President of Huawei Cloud & AI Products & Services, said that Huawei will deploy a three plus one intelligent computing industry strategy based on its five types of proprietary core chips for compute, network, storage, management, and AI. Huawei aims to address the computing requirements in the device-edge-cloud scenarios by leveraging three platforms: the x86, AI, and ARM computing platforms. The three platforms are the foundation for Huawei's full-stack, all-scenario AI solutions. Meanwhile, Huawei will build an intelligent computing ecosystem around the three platforms, working with partners to lay the cornerstones for customers' digital and intelligent transformation.

"AI as a new general-purpose technology will become a new basic capability to change every industry and organization," said Ma Haixu, Director of the Huawei Intelligent Computing Business Department. "Huawei has always advocated providing inclusive AI that is affordable, effective, and reliable. As a vital part of Huawei's AI blueprint, the Atlas AI computing platform has been applied in smart city, carrier, finance, Internet, and electric power industries. Huawei has jointly innovated with partners to launch AI-facing best practices, empowering customers with high-performance, all-scenario industry solutions to embrace pervasive intelligence."