After a decade, Arm Holdings has announced its first new chip architecture, Armv9. The Armv8, released in 2011, has powered some of the best Chromebooks and even the best Android devices. And now, with this new release, Arm expects a big performance improvement of up to 30% over the next two chip generations, but the main focus for this new design is security and AI.
In this modern era, security is becoming increasingly important as more of our lives are put into the digital world. To address this situation, Arm is introducing its Confidential Compute Architecture (CCA), which is a hardware-based secure environment that protects certain data while it’s in use. The way it works is it places the data in “Realms,” which are created dynamically and separates the data from the rest of the system.
For AI, Arm created the Scalable Vector Extension 2 (SVE2) to boost on-device machine learning and digital signal processing. What this will do is it will help CPUs locally process 5G systems and augmented and virtual reality. Furthermore, Arm plans to increase AI capabilities over the next few years. The initial version of this technology was created with Fujitsu to power the world’s fastest supercomputer. Now it will soon be used to power more than 8 billion AI-capable by the mid-2020s.
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.Richard Grisenthwaite (Senior Vice President and Chief Architect)
Powerful chips like the Qualcomm Snapdragon 888 already bring impressive performance to smartphones like the OnePlus 9 Pro, and the Apple M1 chip has proven to be a powerhouse on laptops like the Macbook Air M1. These chips prove the fact that Arm’s architecture is becoming increasingly important for powerfully efficient computing, and the Armv9 design will continue to bring unprecedented performance in “the next 300 billion Arm-based chips.”