News

Armv9 Architecture Addresses Growing Demands of AI

• By John K. Waters
• 03/30/2021

Arm, the chip designer behind the silicon in billions of mobile devices, today unveiled a new chip architecture aimed at meeting "the global demand for ubiquitous specialized processing with increasingly capable security and artificial intelligence (AI)."

Armv9 is the first new Arm architecture to come from the Cambridge, UK-based company in a decade. It builds on the success of its enormously popular Armv8 design.

Among other things, the Armv9 design introduces the Confidential Compute Architecture (CCA). Confidential computing shields portions of code and data from access or modification while in-use, even from privileged software, the company says, by performing computation in a hardware-based secure environment.

The CCA also introduces a concept called "realms," which refers to dynamically created, secured, containerized execution environments useable by all applications. Realms are regions that are separate from both the secure and non-secure worlds.

The new capabilities in Armv9 will accelerate the industry-wide move from general-purpose to more specialized computing across every application as AI, the Internet of Things (IoT), and 5G gain momentum globally, the company claims.

Arm partnered with Fujitsu specifically to address the increased demands of AI in this chip design. Arm built on Fujitsu's work on Scalable Vector Extension (SVE), the tech at the heart of Fugaku, the world’s fastest supercomputer. The result is the new SVE2 for Armv9, which enables enhanced machine learning (ML) and digital signal processing (DSP) capabilities across a wider range of applications.

SVE2 enhances the processing ability of 5G systems, virtual and augmented reality, and ML workloads running locally on CPUs, such as image processing and smart home applications. Over the next few years, the company says, Arm will further extend the AI capabilities of its technology with substantial enhancements in matrix multiplication within the CPU, in addition to ongoing AI innovations in its Mali GPUs and Ethos NPUs.

"Addressing the demand for more complex AI-based workloads is driving the need for more secure and specialized processing, which will be the key to unlocking new markets and opportunities," said Richard Grisenthwaite, SVP, chief architect, and fellow at Arm, in a statement. "Armv9 will enable developers to build and program the trusted compute platforms of tomorrow by bridging critical gaps between hardware and software, while enabling the standardization to help our partners balance faster time-to-market and cost control alongside the ability to create their own unique solutions."

Arm sells processor designs and licenses code to companies like Qualcomm, Samsung, and Apple. Last year, Nvidia agreed to acquire the SoftBank-owned company for $40B.

"Nvidia sees enormous opportunities to bring the transformative powers of AI deeper into gaming, autonomous vehicles, enterprise data centers, and embedded devices," said Brian Kelleher, SVP Nvidia's hardware engineering group, in a statement. "Through our ongoing collaboration with Arm, we look forward to using Armv9 to deliver a wide range of once unimaginable computing possibilities."