Accelerating Ubiquitous Intelligence: Arm discusses requirements of complex AI and ML workloads

 Rene Haas, President of Arm’s IP Products Group (IPG), addressed the topic of “Accelerating Ubiquitous Intelligence” in his keynote at the 2021 COMPUTEX Forum—the annual Taipei computing conference. He starts off the topic with Arm’s involvement in supplying the world with solutions based on the company’s design and in the near future, all digitally shared data will be securely processed, transmitted, and stored from the start until the finish line. Aside from recapping the Armv9’s major architecture and features from the 1st keynote, he also shared the new era of computing is approaching where AI and ML will be the key to every compute environment enabled by specialized processing in which Armv9 is designed to do so from the beginning where he used the flagship Cortex-X2 as an example that brings high CPU and GPU performance while raising the longevity of battery life, a next-generation improvement if you will.

Contributing to specialized processing also requires the software part to be optimized, dynamic, and richer than before. Therefore, Haas introduced the Total Compute solutions that span across tools, IPs, and applications that help partners to develop mature products from the more consumer TV, smartphones, and laptops down to industry use HPC, cloud computing, etc. He also elaborated on the recent announcement of NVIDIA’s Grace CPU that is fully dedicated to AI and ML usage alongside several HPC developer kits.

Lastly, Haas rounded up everything in a simple message: No devices should be left behind, even the smaller and cost-sensitive components can be made smarter and more capable to serve all data at the closest proximity and Armv9 is the key to unlocking the limitless potential of AI and ML.

Interested in finding out all the technical details? Check out the full keynote below.

https://youtu.be/2Smn6MKJCfE