Nvidia unveils Jetson AGX Xavier module for industrial AI deployments

The rugged system-on-module features 512-core Volta GPU with 64 Tensor Cores, two deep learning accelerators, Arm CPU, and more

Nvidia has launched a new compute module designed for AI systems in harsh, safety-critical environments.

Nvidia’s Jetson AGX Xavier Industrial features up to four times more memory than the Jetson TX2i and is claimed to offer 20 times the performance, delivering up to 30 trillion operations per second (TOPS).

The Jetson AGX Xavier Industrial could allow for autonomous deployments in various industrial settings; construction, automobile manufacturing, and die-casting were listed as some of the examples.

Safe and secure

Nvidia has revealed the new Xavier Industrial module three years after it launched the Jetson AGX Xavier as a standalone option.

In 2019, the company unveiled the Jetson Xavier NX, a compact AI ‘supercomputer’ about the size of a credit card designed specifically for robotic and embedded computing devices at the edge.

Nvidia’s latest offering houses a 512-core Volta GPU with 64 Tensor Cores, two deep learning accelerators, two computer vision accelerators, an eight-core Nvidia Carmel Arm CPU, and both an encoder and decoder.

The module also contains a new Safety Cluster Engine (SCE) which features dual Arm Cortex-R5 processors for integrated fault-detection mechanisms, lock-step subsystems, and built-in-system tests.

“Jetson AGX Xavier Industrial combines the supercomputing capabilities of the Jetson AGX Xavier system-on-module with new reliability, availability and serviceability features required to deploy AI in tough environments,” Nvidia’s announcement reads.

“These include error correction codes, single error correction, double error detection, and parity protection to deliver internal RAM resilience, address and data bus error detection and correction and IP resiliency in industrial applications.”

The Jetson AGX Xavier Industrial is available to order now and will be shipping in late July.

Developers can start building embedded systems for their industrial applications today using the Jetson AGX Xavier Developer Kit, downloading JetPack, and designing with all the documentation available on the Jetson site.

Nvidia’s Xavier announcement comes just a few weeks after it unveiled Base Command, a cloud-based development hub for large-scale, multi-team AI projects.

Nvidia also announced that “dozens” of new servers have been certified to run its AI Enterprise software – while teasing that certification efforts will expand to Arm CPU servers in 2022.