Prominent semiconductor designer Arm, which licenses its designs to others, debuted Armv9 architecture, which features its first major architectural advance in a decade and includes a 30 percent improvement in speed. Arm chief executive Simon Segars revealed that Armv9 will be “the base for the next 300 billion Arm-based chips.” Its customers have already shipped 180+ billion chips that impact 70+ percent of the global population. Nvidia is in the process of acquiring Arm, which is based in the United Kingdom, for $40 billion.
VentureBeat reports that, according to Segars, “the new architecture has processing that balances economics, design freedom, and accessibility advantages of general-purpose computing devices with specialized processors that handle tasks like digital signal processing and machine learning.”
He added that, “at the current rate, 100 percent of the world’s shared data will soon be processed on Arm, either at the endpoint, in the data networks or the cloud.” Armv9 chips will also be more specialized to accommodate AI, IoT and 5G applications.
Arm also introduced its Arm Confidential Compute Architecture (CCA), which “shields portions of code and data from access or modification while in use, even from privileged software, by performing computation in a hardware-based secure environment.” It also debuted “the concept of dynamically created Realms, usable by all applications, in a region that is separate from both the secure and non-secure worlds.”
Segars added that Realms, similar to software containers that isolate code but with hardware support, for example, “can protect commercially sensitive data and code from the rest of the system while it is in use, at rest, and in transit.” In a Pulse survey, 90+ percent of enterprise executives queried said that, “if confidential computing were available, the cost of security could come down,” leading to more investment in innovation.
With regard to AI, Arm also inked a partnership with Fujitsu to “create the Scalable Vector Extension (SVE) technology, which is at the heart of Fugaku, the world’s fastest supercomputer.” Out of that, Arm developed “SVE2 for Armv9 to enable enhanced machine learning (ML) and digital signal processing (DSP) capabilities across a wider range of applications.”
With regard to performance improvements, Segars said that its “Total Compute design methodology will accelerate overall compute performance through focused system-level hardware and software optimizations and increases in use-case performance.” The company will apply these “design principles across its entire IP portfolio of automotive, client, infrastructure, and IoT solutions.”
CNET reports that, according to Arm IP Product Group president Rene Haas, the Arm Confidential Compute Architecture will make cloud computing more secure so that, for example, Amazon won’t be able to see the details of a customer’s transactions. Haas said the 30 percent improvement in performance will take place over two years. There are some high-profile holdouts that prefer to “create their own Arm designs.”
Qualcomm chief executive Cristiano Amon didn’t think Arm’s designs were fast enough, saying the “Arm roadmap does not allow us to lead in the CPU performance for the next-generation computing devices.” Arm has also had trouble breaking into the server processor sector, although its Neoverse designs “have gained a foothold with Amazon’s Graviton2 processors.”