Arm and Nvidia Tap Into the Potential of IoT and AI Together

April 2, 2018 - 4 minutes read

IoT app developer, Internet of Things app development

Both artificial intelligence (AI) and the Internet of Things (IoT) appear boundless in potential. The development of these two technologies has spurred immense innovation in many industries already. But combining them could yield greater results than we can even imagine.

Arm and Nvidia are teaming up to make it easier for IoT companies to integrate AI into their products.

Streamlining the Future

Although Nvidia is located in Santa Clara, California, and Arm is headquartered an hour north of London, geographical limitations won’t stop these two tech giants from bringing the future to us faster.

They’re working together to integrate Nvdla (pronounced “enn-vid-lah”), Nvidia’s Deep Learning Accelerator architecture, into Arm’s machine learning platform, Project Trillium. This move would make it much simpler for IoT developers to incorporate AI into their chip designs and, as the two companies put it, build inference into devices. This, in turn, should lead to smarter products becoming more affordable.

While speaking at the annual GPU technology conference, Deepu Talla, the Vice President and General Manager of Autonomous Machines at Nvidia, touched upon the importance of accelerating adoption of these technologies: “Our partnership with Arm will help drive this wave of adoption by making it easy for hundreds of chip companies to incorporate deep learning technology.”

He was extremely optimistic about Nvidia’s partnership with the renowned semiconductor company: “Today we are one step closer to that vision by incorporating Nvdla into the Arm Project Trillium platform, as our entire ecosystem will immediately benefit from the expertise and capabilities our two companies bring in AI and IoT.”

Cutting-Edge Computing

Nvidia’s Xavier, billed by the company as “the Nvidia AI Supercomputer for the Future of Autonomous Transportation,” serves as the basis for Nvdla. The company also supports its deep learning architecture with developer tools like TensorRT, a deep learning accelerator that’s programmable. But Nvdla’s real strength comes from its open-source design. This allows for cutting-edge contributions and new discoveries to get added regularly.

One of the main focuses of the collaboration will be on edge computing. This is when part of the information processing is done on the IoT devices themselves. IoT networks naturally are quite fragile in terms of connectivity and security. Processing mountains of IoT sensor data in real time is also a sizeable problem.

Emphasizing edge computing could solve this. Rene Haas, Executive Vice-President at Arm, believes it’s integral: “Accelerating AI at the edge is critical in enabling Arm’s vision of connecting a trillion IoT devices.”

Deep Potential

The two companies are hopeful that this team-up could result in deep learning eventually being integrated into IoT devices. They believe that the collaboration will give those in the deep learning field unparalleled levels of performance capabilities. “This is a win/win for IoT, mobile, and embedded chip companies looking to design accelerate AI inferencing solutions,” explains Karl Freund, a lead deep learning analyst at Moor Insights and Strategy.

This partnership will undoubtedly affect the adoption of AI and IoT on a deep level. If Nvidia and Arm are successful, expect to be seeing smart IoT devices pop up everywhere in the next few years.

Tags: , , , , , , , , , , , , , , ,