It should be the foundation for “the next 300 billion Arm-based chips.”
Arm has unveiled the Arm v9 architecture, which the developer says is “a response to the global demand for ubiquitous, dedicated processing with enhanced security and artificial intelligence.”
Arm v9 architecture introduced
“As we look to the future that will shape AI, we must lay the foundation for cutting-edge computing that is ready to tackle the unique challenges ahead,” said Simon Segars, CEO, Arm. – Armv9 is the answer. It will be at the forefront of the next 300 billion Arm-based chips driven by the demand for ubiquitous dedicated, secure and powerful processing based on economics, design freedom, and the availability of universal computing. ”
The number of Arm-based chips in the world continues to grow, with more than 100 billion units shipped over the past five years, the company said. At the current rate, 100% of the world’s shared data will soon be processed on Arm, either at the endpoint, data networks, or in the cloud. Armv9’s new capabilities are said to accelerate the transition from universal computing to more specialized computing across all applications as AI, Internet of Things (IoT), and 5G gain traction globally.
The Arm Confidential Computing Architecture (CCA) is designed to keep data secure. Confidential Computing protects portions of code and data from being accessed or modified during use, even from privileged software, by performing computations in a secure hardware environment. At the heart of the Arm, CCA concept are dynamically created Realms used by all applications in a region that is separate from both safe and insecure worlds.
The acceleration of AI algorithms is based on the Scalable Vector Extension (SVE) technology co-developed with Fujitsu, which is at the heart of Fugaku, the world’s fastest supercomputer. Armv9 includes the second generation of this technology, SVE2, which should provide advanced machine learning and digital signal processing capabilities for a wide range of applications.
In terms of performance, Armv9 is expected to increase performance by over 30% in the next two generations of mobile and infrastructure processors.
Moreover, the additional gain should be provided by the Arm Total Compute design methodology, based on “targeted optimization of hardware and software at the system level.” By applying Total Compute design principles, Arm expects to improve the performance of IP cores for automotive, consumer electronics, infrastructure, and IoT solutions. In addition, Arm is developing several technologies to increase the frequency, bandwidth, and cache size, as well as reduce memory latency, to further improve the performance of Armv9-based processors.
