Smallest AI supercomputer for embedded, edge applications

November 09, 2021 // By Rich Pell
Smallest AI supercomputer for embedded, edge applications
Graphics processing chips specialist Nvidia has announced what it says is the world’s smallest, most powerful and energy-efficient AI supercomputer for robotics, autonomous machines, medical devices and other forms of embedded computing at the edge.

Built on the NVIDIA Ampere architecture, Jetson AGX Orin provides 6x the processing power of its predecessor - the Jetson AGX Xavier - while maintaining its form factor and pin compatibility. It delivers 200 trillion operations per second, similar to that of a GPU-enabled server but in a size that fits in the palm of a hand.

The new Jetson computer, says the company, accelerates the full NVIDIA AI software stack, allowing developers to deploy the largest, most complex models needed to solve edge AI and robotics challenges in natural language understanding, 3D perception, multisensor fusion and more.

“As robotics and embedded computing transform manufacturing, healthcare, retail, transportation, smart cities and other essential sectors of the economy, the demand for processing continues to surge,” says Deepu Talla, vice president and general manager of embedded and edge computing at NVIDIA. “Jetson AGX Orin addresses this need, enabling the 850,000 Jetson developers and over 6,000 companies building commercial products on it to create and deploy autonomous machines and edge AI applications that once seemed impossible.”

Jetson AGX Orin features an NVIDIA Ampere architecture GPU and Arm Cortex-A78AE CPUs, along with next-generation deep learning and vision accelerators. High-speed interfaces, faster memory bandwidth and multimodal sensor support provide the ability to feed multiple concurrent AI application pipelines.

Users of the Jetson AGX Orin can leverage the NVIDIA CUDA-X accelerated computing stack, NVIDIA JetPack SDK and the latest NVIDIA tools for application development and optimization, including cloud-native development workflows. Pretrained models from the NVIDIA NGC catalog are optimized and ready for fine-tuning with the NVIDIA TAO toolkit and customer datasets, reducing time and cost for production-quality AI deployments, while cloud-native technologies allow seamless updates throughout a product’s lifetime.

For specific use cases, software frameworks include NVIDIA Isaac Sim on Omniverse for robotics, NVIDIA Clara Holoscan SDK for healthcare, and NVIDIA DRIVE for autonomous driving. The latest Isaac release includes significant support for the Robot Operating


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.